0% found this document useful (0 votes)
159 views208 pages

The Cybersecurity Mesh Architecture (2025) by Kumar T

The document discusses the Cybersecurity Mesh Architecture (CSMA), a flexible and scalable security approach aimed at creating a resilient security ecosystem. It outlines the evolution of cybersecurity, current challenges, and the benefits of adopting CSMA, emphasizing the need for integration and collaboration among security tools. The book serves as a resource for students and technical readers to enhance their understanding of CSMA and its application in modern cybersecurity strategies.

Uploaded by

Lawrence Leong
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
159 views208 pages

The Cybersecurity Mesh Architecture (2025) by Kumar T

The document discusses the Cybersecurity Mesh Architecture (CSMA), a flexible and scalable security approach aimed at creating a resilient security ecosystem. It outlines the evolution of cybersecurity, current challenges, and the benefits of adopting CSMA, emphasizing the need for integration and collaboration among security tools. The book serves as a resource for students and technical readers to enhance their understanding of CSMA and its application in modern cybersecurity strategies.

Uploaded by

Lawrence Leong
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 208

The Cybersecurity

Mesh
Architecture
Composable, flexible, and scalable
security approach
for a resilient security ecosystem

Tarun Kumar

www.bpbonline.com
First Edition 2025
Copyright © BPB Publications, India
ISBN: 978-93-65891-218

All Rights Reserved. No part of this publication may be reproduced, distributed or transmitted in any
form or by any means or stored in a database or retrieval system, without the prior written permission
of the publisher with the exception to the program listings which may be entered, stored and executed
in a computer system, but they can not be reproduced by the means of publication, photocopy,
recording, or by any electronic and mechanical means.

LIMITS OF LIABILITY AND DISCLAIMER OF WARRANTY


The information contained in this book is true to correct and the best of author’s and publisher’s
knowledge. The author has made every effort to ensure the accuracy of these publications, but
publisher cannot be held responsible for any loss or damage arising from any information in this
book.
All trademarks referred to in the book are acknowledged as properties of their respective owners but
BPB Publications cannot guarantee the accuracy of this information.

www.bpbonline.com
Dedicated to
My beloved wife:
Kanchan Bhatia Kumar
and
My son Vihaan Kumar
About the Author

Tarun is a seasoned professional with 25+ years of exclusive experience


covering Cyber/Information Security, IT Risk Management, Data
Protection, and Privacy. In various leadership roles, he has had the
distinction of building cybersecurity capabilities (people, process, and
technology) for organizations (with extensive experience in managing large
teams).
He has been CISO with a large global automobile major. In his past roles,
he has been Director of Cybersecurity with a Big 4 firm; Global CISO with
a leading global Business Process Management (BPM) organization and
General Manager of Global Security Services with a large global IT
Services organization.
He has extensively been involved in providing Cybersecurity services to
domestic (India) and international clients across industries. He brings a 360-
degree experience in Cybersecurity as a provider (consulting) and consumer
(CISO) of services.
He has extensive experience in establishing road maps and investments for
Cybersecurity, security governance, and IT risk management practices. His
proficiency covers Cyber Security strategy and roadmap, governance, risk,
compliance and controls, threat and vulnerability management, operations,
data privacy, and business continuity.
He has been recognized with various industry honors between 2012-2023.
He brings recognized credentials of CRISC, CISM, CISA, CISSP, and
ISO27001 Lead Auditor.
About the Reviewer

John Mathew is a Cybersecurity Architect with eight years of experience


in the field. He specializes in Penetration Testing, Ethical Hacking, Red
Teaming and Blue Teaming. John holds a Bachelor of Technology in
Computer Science and Engineering and has earned notable accolades,
including First place in Capture the Flag contests and certifications such as
ISC Certified in Cybersecurity and EC-Council Certified Ethical Hacker.
Outside of work, he is an avid reader and enjoys playing video games. He
believes that Security is a myth and Testing is inevitable, a philosophy that
drives his commitment to excellence in Cybersecurity.
Acknowledgement

I want to express my deepest gratitude to my family and friends for their


unwavering support and encouragement throughout this book's writing,
especially my wife Kanchan and son Vihaan.
I am also grateful to BPB Publications for their guidance and expertise in
bringing this book to fruition. Revising this book was a long journey, with
valuable participation and collaboration of reviewers, technical experts, and
editors.
I have always thought that what makes the job worthwhile are the people
you work with. I am thankful for the people I have worked with – on my
team, across departments, and in my ecosystem, and would like to
acknowledge their valuable contributions.
I would also like to thank my mentors, stakeholders, and colleagues for
their continuous support and guidance.
Finally, I would like to thank all the readers who have taken an interest in
my book and for their support in making it a reality. Your encouragement
has been invaluable.
Preface

We reside in a world where the field of cyber threats is enormous and ever-
evolving. Every new security solution/tool seems to result in newer ways
for attackers to circumvent defenses.
The one standout pluck from the book is that a robust cybersecurity posture
in today’s times necessitates the amalgamation of and partnership
(collaboration) between the various security solutions/tools that have been
deployed.
Contemporary technology stacks are extensively distributed and often
difficult to manage when separated into individual Silos. Hence, partnership
(collaboration), integration, and aggregation are critical features of a
successful cybersecurity strategy.
The book explores the concept of Cybersecurity Mesh Architecture
(CSMA). After reading through all chapters, readers will appreciate the fact
that CSMA is a valuable asset to enterprises (businesses) since it is an
architectural philosophy that advocates solution/tool integration and data
aggregation to achieve the desired outcomes. It also provisions for security
analytics, integrated threat intelligence/dashboards, and automation
supported by AI to achieve a cybersecurity posture that is dynamic and
capable of responding swifter than attackers.
This book is suitable for students who are studying cybersecurity as a
subject in their bachelor/master programs. It is also written for technical
readers with a basic understanding of cybersecurity and networking
technologies and their challenges.
This book is a resource that will enable you to have more trust in your
knowledge of CSMA. I hope you will find this book informative and
helpful.
Chapter 1: Cybersecurity: A Dynamic Changing Paradigm – This
chapter reviews the chronology of the evolution of cybersecurity, presents a
detailed overview of some noteworthy cybersecurity events (2010 – to
date), takes a look at some major trends that had a noteworthy impact on
cybersecurity, and examines the building blocks of cybersecurity and
traditional cybersecurity measures.
Chapter 2: Cybersecurity: Understanding Today’s Security Challenges
– This chapter covers topics such as distributed systems, examines the
security challenges of distributed systems, and presents details about
cybersecurity threats, attacks, and key issues in the digital age.
Chapter 3: Emerging Cybersecurity Trends – In this chapter, we will
explore the cybersecurity trends of today and the future, concentrating on
presenting the common themes in these trends. This chapter also allows the
reader to understand the importance of cyber resilience.
Chapter 4: The Need for Cybersecurity Mesh Architecture –This
chapter presents the current situation of the cybersecurity ecosystem,
explains CSMA, and illustrates its layers, needs, and benefits.
Chapter 5: Fundamental Components of Cybersecurity Mesh
Architecture – This chapter gives special attention to the key components
of CSMA, discusses the outcome of the adoption of CSMA, a unified
architecture and provides a sneak preview of CSMA products/solutions.
Chapter 6: How to Effectively Adopt Cybersecurity Mesh Architecture
– This chapter reassesses the cybersecurity landscape of today, elaborates
on the key aspects of CSMA adoption, provides directions on how to get
started with CSMA, and discusses the key factors of consideration while
adopting CSMA.
Chapter 7: Benefits of Adopting Cybersecurity Mesh Architecture –
This chapter emphasizes the necessity of CSMA and the benefits of
leveraging CSMA. The chapter then discusses the characteristics of a
CSMA strategy and presents a few target use cases. Furthermore, it details
the features to be considered for CSMA solutions and presents the pitfalls
of not leveraging CSMA.
Chapter 8: CSMA Best Practices – In this chapter, we will compare
CSMA with the traditional defense-in-depth approach and re-visit the
salient points and goals. We will also discuss a systematic approach to
implementing CSMA and take a look at the KPIs for assessing the
effectiveness of the implementation of CSMA. The chapter also covers the
commandments of CSMA and discusses the challenges in implementing
CSMA.
Chapter 9: Potential Outlook for CSMA Adoption – This chapter will
cover three distinct use cases in different environments where CSMA works
[viz., work from home, cloud, and operational technology (OT)]. The
chapter will also examine the use of CSMA in the healthcare sector and
take a look at the CSMA market overview, its growth factors, dynamics,
and growth opportunities.
Coloured Images
Please follow the link to download the
Coloured Images of the book:

https://rebrand.ly/75e90aj
We have code bundles from our rich catalogue of books and videos
available at https://github.com/bpbpublications. Check them out!

Errata
We take immense pride in our work at BPB Publications and follow best
practices to ensure the accuracy of our content to provide with an indulging
reading experience to our subscribers. Our readers are our mirrors, and we
use their inputs to reflect and improve upon human errors, if any, that may
have occurred during the publishing processes involved. To let us maintain
the quality and help us reach out to any readers who might be having
difficulties due to any unforeseen errors, please write to us at :
[email protected]
Your support, suggestions and feedbacks are highly appreciated by the BPB
Publications’ Family.
Did you know that BPB offers eBook versions of every book published, with PDF and ePub files
available? You can upgrade to the eBook version at www.bpbonline.com and as a print book
customer, you are entitled to a discount on the eBook copy. Get in touch with us at :
[email protected] for more details.
At www.bpbonline.com, you can also read a collection of free technical articles, sign up for a range
of free newsletters, and receive exclusive discounts and offers on BPB books and eBooks.

Piracy
If you come across any illegal copies of our works in any form on the internet, we would be grateful
if you would provide us with the location address or website name. Please contact us at
[email protected] with a link to the material.

If you are interested in becoming an author


If there is a topic that you have expertise in, and you are interested in either writing or contributing to
a book, please visit www.bpbonline.com. We have worked with thousands of developers and tech
professionals, just like you, to help them share their insights with the global tech community. You can
make a general application, apply for a specific hot topic that we are recruiting an author for, or
submit your own idea.

Reviews
Please leave a review. Once you have read and used this book, why not leave a review on the site that
you purchased it from? Potential readers can then see and use your unbiased opinion to make
purchase decisions. We at BPB can understand what you think about our products, and our authors
can see your feedback on their book. Thank you!
For more information about BPB, please visit www.bpbonline.com.

Join our book’s Discord space


Join the book’s Discord Workspace for Latest updates, Offers, Tech
happenings around the world, New Release and Sessions with the Authors:
https://discord.bpbonline.com
Table of Contents

1. 1. Cybersecurity: A Dynamic Changing Paradigm


1. Introduction
2. Structure
3. Objectives
4. Evolution of cybersecurity
1. Time before networking
2. Time of phone phreaks
3. Time of early hacking
4. Time of ARPANET and the Creeper and
Reaper
5. Decade of commercial antivirus
1. The first DoS attack
6. Connected world
1. The time of email
7. Cybersecurity threat canvas expansion
1. The Anonymous
2. Operation Aurora
3. Stuxnet
4. General Data Protection
Regulation
5. The X hack
6. The SolarWinds attack
7. Remote and hybrid work
8. Use of artificial intelligence in
cybersecurity
5. Notable cybersecurity events
1. Events in 2013
2. Events in 2014
3. Events in 2015 and 2016
4. Events in 2017
5. Events in 2018
6. Events in 2019
7. Events in 2020
8. 2021-2023 and beyond
6. Notable shifts impacting cybersecurity
7. Cybersecurity threats evolution
8. Building blocks of cybersecurity
9. Traditional cybersecurity measures
10. Conclusion
11. Points to remember
12. Key terms
13. References

2. 2. Cybersecurity: Understanding Today’s Security Challenges


1. Introduction
2. Structure
3. Objectives
4. Distributed systems
1. Relatively easy to permeate
5. Security challenges of distributed systems
1. Lack visibility
2. Difficult to manage
3. Digital age: Cybersecurity threats
1. Sources of cyber threats
6. Digital age: Cybersecurity attacks
1. Malware attacks
1. Ransomware attacks
2. Cryptojacking attacks
2. Supply chain attacks
3. Social engineering attacks
1. Phishing attacks
2. Generative artificial
intelligence phishing attacks
4. Internet of Things attacks
1. Smart medical devices and
Electronic Medical Records
vulnerabilities
2. Privacy concerns of connected
cars
5. Cloud attacks
6. State-sponsored attacks
7. Insider attacks
8. Man-in-the-middle attack
9. Denial-of-Service attack
10. Injection attacks
7. Digital age: Cybersecurity key issues
1. Budgets issue
2. The skills gap issue
3. Artificial intelligence issues
8. Conclusion
9. Points to remember
10. Key terms
11. References

3. 3. Emerging Cybersecurity Trends


1. Introduction
2. Structure
3. Objectives
4. Cybersecurity trends
1. Trend 1: Tenacious data breaches
2. Trend 2: Upsurging ransomware
3. Trend 3: Smarter social engineering attacks
4. Trend 4: Imminent cloud security threats
5. Trend 5: Threat emanating due to mobile
devices
6. Trend 6: Evolving IoT threats
7. Trend 7: Growing power of AI
8. Trend 8: Intensifying cyber warfare by rogue
nations
9. Trend 9: Improving MFA
10. Trend 10: Enhanced automation
11. Trend 11: Rise of data privacy
12. Trend 12: Risks of distributive workforce
13. Trend 13: Rise in insider threats
14. Trend 14: Widening IT skills gap
15. Trend 15: Rising threats of Deepfakes
16. Trend 16: More supply chain attacks
17. Trend 17: Continued prominence of
regulations
18. Trend 18: Emphasis on incident response and
recovery
19. Trend 19: Cyber resilience
20. Trend 20: A board room agenda
5. Cyber security future trends
1. Future trend 1: Zero trust
2. Future trend 2: Cryptography resistant to
quantum computing
3. Future trend 3: 5G network security
4. Future trend 4: Stronger authentication
measures
5. Future trend 5: Human-centric security
6. Future trend 6: cyber insurance
7. Future trend 7: AI-assisted security testing
8. Future trend 8: AI-driven cybercrime
9. Future trend 9: Smart cities and critical
infrastructure security
10. Future trend 10: Ethical hacking and bug
bounty programs
6. Common themes in trends
7. Understanding cyber resilience
1. Importance of cyber resilience
2. Building blocks of cyber resilience
8. Conclusion
9. Points to remember
10. Key terms

4. 4. The Need for Cybersecurity Mesh Architecture


1. Introduction
2. Structure
3. Objectives
4. The current situation of cybersecurity ecosystem
5. CSMA explained
6. Illustrating the layers of CSMA
7. The need for CSMA
1. Challenge 1: Data deluge
2. Challenge 2: Data analysis
3. Challenge 3: Real-time security monitoring
4. Challenge 4: Attackers think holistically
5. Challenge 5: Disjointed perimeter
6. Challenge 6: Adoption of multi-cloud strategy
8. Benefits of CSMA
1. Enhancement of visibility
2. Reduction in complexity
3. Enhanced sharing of intelligence
4. Enhanced automation
5. Consolidation of security products
6. Security analytics
9. Conclusion
10. Points to remember
11. Key terms

5. 5. Fundamental Components of Cybersecurity Mesh Architecture


1. Introduction
2. Structure
3. Objectives
4. A re-look at CSMA
1. CSMA working relationship with zero trust
2. CSMA’s scalability relative to other
architectures
3. Working of CSMA
5. Key components of CSMA
1. Integrated data
2. Distributed identity fabric
3. Comprehensive security intelligence
4. Automated operations
5. Centralized operations
6. Consolidated dashboards
1. CSMA’s most important
component
6. The unified architecture of CSMA
1. Extensive
2. Integrated
3. Accelerated through automation
7. Major products/solutions of CSMA
8. Conclusion
9. Points to remember
10. Key terms

6. 6. How to Effectively Adopt Cybersecurity Mesh Architecture


1. Introduction
2. Structure
3. Objectives
4. Performance
1. Problem with past solutions
2. Answer lies in the unified architecture
5. Cybersecurity landscape of today
1. Attack surface expansion
2. Adoption of cloud
6. Key aspects for CSMA adoption
1. Integration
2. Comprehensiveness
3. Automation
7. Getting started with CSMA
1. Assess current tools and identify gaps
2. Build the key components
3. Evaluate compatible tools
4. Consider befitting vendor(s)
5. Seek consolidation
6. Review and revise
8. Key factors to consider while adopting CSMA
9. Conclusion
10. Points to remember
11. Key terms

7. 7. Benefits of Adopting Cybersecurity Mesh Architecture


1. Introduction
2. Structure
3. Objectives
4. What is the necessity of CSMA
5. Benefits of leveraging CSMA
1. Enhanced agility and flexibility
2. Enhanced security intelligence
3. Enhanced incident response
4. Efficient compliance
5. Enhanced savings
6. Characteristics of a CSMA strategy
7. Target use cases
1. Security operations
2. Vulnerability management
3. Compliance management
8. Features to be considered for CSMA solutions
1. API Graph
2. Data normalization and correlation
3. Natural language processing
4. Generative AI for context enhancement
9. Pitfalls of not leveraging CSMA
1. Increased security overheads
2. Increased security risk
3. Inefficient security incident management
4. Restricted decision-making
10. Conclusion
11. Points to remember
12. Key terms

8. 8. CSMA Best Practices


1. Introduction
2. Structure
3. Objectives
4. CSMA vs. defense-in-depth approach
5. Salient points and goals of CSMA
1. CSMA needs to be vendor-agnostic
6. Systematic approach to implementing CSMA
1. CSMA standouts (in comparison to other
architectures)
7. CSMA key performance indicators
8. The ten commandments of CSMA
9. Challenges in implementing CSMA
10. Conclusion
11. Points to remember
12. Key terms

9. 9. Potential Outlook for CSMA Adoption


1. Introduction
2. Structure
3. Objectives
4. Work-from-home environment
1. Zero-trust network access
2. Endpoint detection and response
3. Home network security
5. OT security environment
6. CSMA in the healthcare industry
1. Summarizing CSMA in healthcare
7. CSMA
1. Global market overview and growth factors
2. Market dynamics
1. Driver: Management of
evolving threat landscape
2. Threat: Shortage of skilled
expertise
3. Opportunity: Growing
adoption of multi-cloud
8. Conclusion
9. Points to remember
10. Key terms
11. Multiple choice questions
1. Answer key

10. Index
CHAPTER 1
Cybersecurity: A Dynamic Changing
Paradigm

Introduction
Not only have the domains of cybersecurity and technology advanced but
also have criminals/bad actors who aim to exploit weaknesses in the system
for personal gain. Likewise, cybersecurity and cybercrime have
progressively developed from the 1940s to the present, and this chapter
explains the evolution of cyberattacks and security solutions.

Structure
In this chapter, we will cover the following topics:

Evolution of cybersecurity
Notable cybersecurity events
Notable shifts impacting cybersecurity
Cybersecurity threats evolution
Building blocks of cybersecurity
Traditional cybersecurity measures

Objectives
When we are asked when cybersecurity started, an instant answer, in most
cases, is when the Internet started. Essentially, in this chapter, we shall
realize that the cybersecurity industry has been growing since the 1940s.
Even when networks did not exist, theorists were getting prepared for the
risks that may emerge with the advancement of technology.
In this chapter, we will explore the history and evolution of cybersecurity
—from the age of the first computer threats to the rise of risks due to the
advent of artificial intelligence and cloud computing.

Evolution of cybersecurity
The evolution of cybersecurity has been hand in glove with the
developments in communication technology. Thereafter, its evolution has
been influenced by developments in geopolitical tensions and major global
events.
For the sake of discussion, let us keep our focus on developments in
technology. Over the years, we have progressed from telegraphs to
smartphones. With this, the types of cyberattacks have transformed from
Morris Worm (refer to The first DoS attack section) to Stuxnet (refer to
Stuxnet section for details), to Snake ransomware. Also, to secure data and
communication, we have dealt with simple ciphers to sophisticated
algorithms.
There have been umpteen incidents in our history that have influenced the
developments in cybersecurity. However, it is not feasible to describe all of
them. While we look back in history, some crucial junctures need to be
given their due importance and thus cannot go unnoticed.
Let us take the time capsule and go back to the 1940s.

Time before networking


1943 saw the arrival of the first digital computer. Interestingly, electronic
machines existed which, were not networked. Moreover, only a small group
of people had access, and many did not know how to operate them. Hence,
cyberattacks were challenging to execute because the threat was virtually
non-existent.
It is very intriguing to note that the theory regarding computer viruses first
came to the fore in the late 1940s. This was when the computer pioneer,
John von Neumann, first raised the prospect of computer programs
reproducing themselves.

Time of phone phreaks


In the 1950s, telephone use came into being, and a new phrase, phone
phreaks, was coined. Phone phreaking became prevalent in the late 1950s.
The phrase refers to various practices used by Phreaks or those interested in
the functioning of phones and tampering with their protocols. This allowed
telecom experts to operate remotely on the network and place free calls,
thus avoiding paying long-distance charges. The initial purpose of hacking
was not information gathering, which it is now.
Though phone phreaking practice gradually disappeared in the 1980s (after
creating havoc for almost three decades), phone providers (who can be
compared to today’s enterprises) were powerless to halt the phreaks (who
can be compared to today’s cyber attackers).

Time of early hacking


The 1960s saw the arrival of massive mainframes that were kept in
controlled environments. Even for programmers, access was restricted
because these computer devices were a high expense.
The majority of the development of the phrase hacking took place during
this decade. It was again not the motive of information gathering by which
the hacking took place, rather it was driven by the motive of breaking into
high-tech machines to alter their functionality.
In this decade, the idea of hacking shifted to computers. We would rather
term it early hacking. As discussed, the motive of hacking was not big
business; the motive of these early hacking incidents was to gain access to
systems. Hacking was not performed for political or economic gains, it was
primarily to ascertain if messing up with computers was possible. It not
only remained a new concept, newer, faster, and more innovative, hacking
techniques also started to emerge during this decade.
In 1967, IBM invited some students to see the newly made computer
system. The students were provided training on various system components
of the computer system so that IBM could gather information about the
system's weaknesses.
As a result, the idea and concept of implementing defensive security
measures to prevent hacking came into effect. We can state that IBM
allowed the students to perform ethical hacking (rather, this could be the
first time ethical hacking was performed in the industry).
Nowadays, we all know that ethical hacking has become a highly sought-
after profession (skill). On reflection, big strides in cybersecurity
developments took place in this decade. The years to follow saw an increase
in computer use. The size of computers not only decreased (became
compact), but due to their affordability, enterprises/businesses started to use
them to store data.
Another entrant to the scene was passwords which were used to access and
secure computers.

Time of ARPANET and the Creeper and Reaper


The actual start (and need) of cybersecurity was seen in the 1970s. The
Advanced Research Projects Agency Network (ARPANET), also known
as Early Internet, came up during this time. In fact, ARPANET (the
connectivity) network was built before the Internet was created. 'I'm the
Creeper; catch me if you can!' was printed using a program developed by
an ARPANET developer using PCs connected to the network. For the first
time, this program switched from one machine (mainframe computers) to
another by itself. The Creeper, therefore, is viewed as the world's first
Computer Worm (code that could be transported between machines),
although it was not destructive in nature.
A program called Reaper was created by another ARPANET researcher
which was designed to move through ARPANET deleting copies of Creeper
worm. The Reaper is known as the first anti-virus.

Decade of commercial antivirus


The two terms: Trojan horse (when early malware specimens emerged)
and computer virus made a formal announcement during this time.
We all remember this time of the Cold War between the United States and
the Soviet Union, which led to an increase in the threat of cyber espionage.
In 1985, the Trusted Computer System Evaluation Criteria, also referred to
as The Orange Book, was published by the US Department of Defense,
that guided on:

Assessing the grade of trust to be placed in software that processes


classified/sensitive information.
The security measures needed by manufacturers to be built into their
commercial products.

This is the decade when the history of computer crime took


shape. Consequently, cybersecurity started to be taken more seriously.
Savvy users quickly learned to monitor the command.com file size, having
noticed that an increase in size was the first sign of potential infection.
Although there are claims for the innovator of the first antivirus product,
1987 is widely regarded as the birth year of commercial antivirus because:

The first antivirus was released by German inventors (Andreas Luneng


and Kai Figge).
Version 1.0 of NOD antivirus was created by 3 Czechoslovakians
First VirusScan was released by McAfee (now Intel Security)
Anti4us and Flushot Plus (commercial antivirus programs) were
released.

The first DoS attack


The first DoS attack took place due to an unintentional error in a computer
worm (Morris worm) code. This code was intended to gauge the size of the
Internet. The error caused the worm to replicate continually to the point that
the ARPANET was clogged, and 10% of all the systems that were
connected crashed. Robert T. Morris, the person who created the Morris
worm, was the first person to be charged under the Computer Fraud and
Abuse Act.

Connected world
This decade gave us the growth and development of the Internet (beyond
proportions). The cybersecurity domain expanded along with it. Many
significant developments in computer security took place in this decade,
such as:

Fears of polymorphic viruses emerged since it was difficult to detect.


1990 witnessed a code that altered itself while spreading through
computing systems.
In 1995, the Secure Sockets Layer (SSL) protocol was developed by
Netscape as a way to keep people secure online data and securely use
Internet transactions and browse. Later, it would act as the foundation
for HyperText Transfer Protocol Secure (HTTPS).

The time of email


Towards the end of the 1990s, email was revolutionizing communication
and was growing in leaps and bounds. However, it also opened up a new
entry avenue for viruses.
1999 witnessed the release of the Melissa virus. The virus found its way
into the user’s computer through MS Word and then emailed copies of itself
to the first 50 email addresses in Microsoft Outlook. The damage costed
around $80 million to fix.

Cybersecurity threat canvas expansion


The 2000s witnessed further growth of the Internet. The majority of homes
and enterprises (businesses) now had computers and this presented new
opportunities for cybercriminals. This was the time when infections on
websites and instant messaging systems were on the rise. This decade also
led to an increase in hacks of credit cards (credit card data leaks).
With the proliferation of the Internet across homes and offices globally,
there were more devices and software vulnerabilities for cybercriminals to
exploit than before. As more data were being kept digitally, there was more
to steal:

In 2001, a new technique to infect computer systems appeared where


there was no need for users to download files to get their systems
infected (they were simply required to visit an infected website). Bad
actors could replace clean pages with infected ones or hide malware on
legitimate web pages.
Also, worms designed to propagate via Internet Chat Relay (IRC)
channel had arrived and instant messaging services also began to get
attacked.
The term zero day attack was coined during this time and this was the
time of these attacks were making news. Zero day attack would mean
exploiting holes in security for new applications and software (with an
attack taking place when hackers exploit the flaw before being
addressed by the developers).
The development of zero-day attacks led to antivirus becoming less
effective.
Another innovation in this decade was operating system (OS) security
– Cybersecurity, that is built into the OS, providing an additional layer
of protection. This includes performing regular OS patch updates,
installing updated antivirus and software, firewalls, and secure
accounts with user management.
Smartphones also proliferated, and this required antivirus software to
be developed for Android and Windows mobiles.

The Anonymous
2003 saw the debut of Anonymous, a popular hacktivist group. The
objective of carrying out attacks by the group is to expose high-profile
targets and garner attention regarding its political views.

Operation Aurora
Until now, no one had thought that Cyber operations could be used as a tool
to carry out industrial espionage at a large scale. Operation Aurora, a series
of cyberattacks brought this to the fore. The attack was sponsored by the
Chinese Government and targeted the intellectual property of many U.S.
private-sector companies, including Google, Yahoo, and Adobe.

Stuxnet
Multiple Windows zero-day vulnerabilities were exploited by Stuxnet (a
sophisticated computer worm). It is alleged to have been created by a secret
U.S.-Israeli program to target and destroy centrifuges at the uranium
enrichment facility in Iran, causing extensive damage to its nuclear
program.
Cybercriminals had newer opportunities to exploit owing to the increase in
connectedness and ongoing digitization.

General Data Protection Regulation


The General Data Protection Regulation (GDPR) is a compliance
regulation that provides citizens of the European Union (EU) greater
control over their personal data. Under this directive, organizations are
responsible for safeguarding the personal data and privacy of EU citizens.
The GDPR applies not only to all the organizations within the EU but also
to organizations outside of the EU that offer goods/services to
customers/businesses in the EU. It was approved by the European
Parliament in April 2016 and came into force on May 25, 2018.

The X hack
This year witnessed dramatic cybersecurity incidents that involved X
(formerly Twitter) accounts of prominent users getting hacked.
A new category of the breach was coined with this, viz., Insider Threat.
Whether it is the actions of a negligent employee/or malicious insider, it
went on to showcase that humans are the weakest link in the cybersecurity
chain.

The SolarWinds attack


The above-mentioned X hack was quite dramatic, however a more ill-famed
hack related to supply chain was unfolded in 2020. This was the exploit of
SolarWinds which caused aftershocks globally (the full consequences of the
attack are still largely unknown).
The SolarWinds Supply Chain Attack is a global hack, as threat actors
turned the SolarWinds’ Orion software into a weapon gaining access to
several government systems and thousands of private systems around the
world. The cybercriminals were able to corrupt one of the servers which
provided access to patches and updates for SolarWinds Orion software.

Remote and hybrid work


The times of the COVID-19 pandemic witnessed remote work model being
adopted by most enterprises (businesses), not by design, but in the wake of
the circumstances. Though this transition was not easy for many
organizations, we are still witnessing hybrid work scenarios in place even
after the pandemic has ended. We have seen a rise in the use of work tools
to enable smooth team work remotely. With this, the security aspect has
rapidly evolved well and is expected to continue in the future. Techniques
such as multi-factor authentication (MFA), better encryption techniques,
principles of zero trust, etc. are becoming commonplace.

Use of artificial intelligence in cybersecurity


The aspect of real-time detection of threats and automated incident response
is improving continuously, thanks to the efficient machine learning (ML)
algorithms and integration of AI (rather seamless) in cybersecurity. Cyber-
attacks can be detected in early stages (can also be pre-empted in cases)
since threat correlation engines are becoming more refined (therefore more
effective). These are in effect becoming pivotal defence mechanisms for
enterprises (businesses).
As we look at these milestones in cybersecurity, it becomes apparent that
the threat landscape is constantly evolving. Cyberattacks are not only a
concern for enterprises (businesses) and governments but also a fear for
every individual who is on the Internet. All entities are equally susceptible.
The following figure displays the evolution of cybersecurity along with a
timeline view:
Figure 1.1: Timeline view of evolution of cybersecurity

Notable cybersecurity events


The story of the evolution of cybersecurity will be incomplete without a
brief mention of some of the notable and unique cybersecurity
incidents/breaches that took place between 2013 and most recently
impacting the national security of countries and costing enterprises
(businesses) millions.

Events in 2013
Some of the noteworthy events of 2013 are listed as follows:

Singaporean government: Anonymous (a hacktivist group) launched


cyberattacks on the Singaporean government. These were a series of
assaults in revenge for Singapore's censorship laws for the Internet.
Israeli government: Hacktivists targeted the Israeli government
touting the campaign as #OpIsrael (which was tied with Holocaust
Remembrance Day). This became an annual synchronized cyberattack
(using many methods such as DDoS).
NSA: Edward Snowden, a former CIA employee (US Government),
copied and leaked classified information from the National Security
Agency. The leaked documents were passed on to The Guardian, who
made them public.
Target: Security lapse at Target (US retailer) resulted in the exposure
of the personal information of 70 million consumers (in addition to the
credit card information of tens of millions of shoppers).

Events in 2014
Some of the noteworthy events of 2014 are listed below:

Home Depot: Credit card information of customers of the Home Dept


(US Retailer) was exposed due to specially crafted malware. The
hackers first breached the Home Depot’s network using credentials
that were stolen from a service vendor and then installed the malware
on the retailer's machines (self-checkout): Quantum – 56 million.
Sony Pictures Entertainment: Confidential data of executives
(salaries, employee details, plans for future films, and some film
scripts) was leaked by a hacker group (Guardians of Peace). The
investigation concluded government of North Korea was behind the
attack.

Globally, it is reported that the number of data breaches in 2014 was about
50% more as compared to 2013.

Events in 2015 and 2016


Some of the noteworthy events of 2015 and 2016 are listed below:

Experian data breach: User records at Experian (a global data


analytics and consumer credit reporting company) were compromised.
This was due to a human mistake in a customer verification process:
Quantum – 15 million.
Snapchat data leak: An unnamed hacker/organization had posted the
phone numbers and usernames of Snapchat users online for free:
Quantum ~4.6 million.
Democratic National Committee (DNC) data breach
The DNC computer network was breached by Cozy Bear and
Fancy Bear (part of Russian intelligence agencies).
This case is often covered as cybercrime in politics as it was
alleged to be carried out by Russia in support of Donald Trump
(during the U.S. Presidential election of 2016).

Events in 2017
Some of the noteworthy events of 2017 are listed below:

Equifax breach: A data breach at Equifax (the American credit


reporting agency) exposed the personal details of about 145+ million
people.
The Shadow Brokers leaks: A hacker group (named The Shadow
Brokers) leaked EternalBlue, a hacking tool used by the National
Security Agency. EnterrnalBlue is an exploit that utilizes
vulnerabilities in the Server Message Block protocol of Windows.
WannaCry ransomworm attack: World’s first and most infamous
ransomworm targeted Windows computer systems (globally) infecting
200,000+ computers in one day.
NotPetya ransomware attack: Petya and NotPetya are malware that
encrypt data. In June 2017, NotPetya was used to perpetrate a
cyberattack globally. WannaCry and NotPetya used the EternalBlue
exploit to affect unpatched computers.
Bad Rabbit ransomware attack: Bad Rabbit ransomware spreads
through drive-by attacks. It appeared as an update for Adobe Flash
(masquerading) that tricked users into downloading it. It asked for
$280 in Bitcoin (providing a 40-hour deadline).

Events in 2018
Some of the noteworthy events of 2018 are listed as follows:

Privacy concerns at Facebook: Access tokens (that were stolen) were


used to expose the accounts of millions of individuals on Facebook.
This was ultimately made public by the company.
Marriott cyberattack: Marriott’s reservation system was
compromised. The data breach went unnoticed for 4 years (probably
took place in 2014, however, was hidden), and impacted 500 million
guests of the hotel.
British Airways cyberattack: Hundreds of thousands of records
(personal data of customers and employees) were breached.
California Consumer Privacy Act (CCPA) signed into law: Due to
the increase in complexity and sophistication of cyber threats, CCPA
was signed into law in 2018 and it went into effect in 2020. It included
an array of privacy rights for consumers and obligations for businesses
concerning the collection and sale of personal information.

Events in 2019
Some of the noteworthy events of 2019 are listed as follows:
Singapore’s health sectors’ breaches: 30+ breaches took place in
Singapore’s healthcare sector in 2019 alone. The number increased to
80+ in 2020.
Attacks on the New Zealand stock market: New Zealand’s stock
market came to a grinding halt due to multiple DDoS attacks.

Events in 2020
This was the year of COVID-19 and proved challenging for cybersecurity
professionals. The year witnessed cybercriminals continued illegal
activities. Some of the important data breaches of 2020 are listed below:

Personal information of MGM Resorts Hotels guests was leaked on a


hacking forum: Quantum – 10+ million.
Facebook profiles were available on the dark web for sale: Quantum –
250+ million.
Zoom accounts were available on the dark web for sale: Quantum -
500k+.
Cognizant Technology Solutions was subjected to a ransomware attack
(perpetrated by the Maze group).

2021-2023 and beyond


These years continued to see many cyberattacks. These cyberattacks can be
categorized as data breaches, ransomware, and data extortion attacks and
they have had a broad impact on enterprises (businesses) during this time
and are expected to continue in the future. Amongst these categories of
cyberattacks, the most prevalent risk to any enterprise’s data security is
ransomware. Unfortunately, over these years, the world has witnessed its
use and ransomware attacks have been on the rise.
According to Verizon's Report, 2023, external threat actors are responsible
for the vast majority (83%) of breaches, and financial gain accounts for
almost all (95%) breaches.
The positive development is in the form of cybersecurity markets that have
been expanding fast as well. According to Statista (Global Statistics portal
for Market Data, Market Research and Market Studies), the size of the
cybersecurity market worldwide by 2026 is anticipated to increase to
$345.4 billion.

Notable shifts impacting cybersecurity


So far, we have seen the historical timeline of the evolution of cybersecurity
and some of the major cybersecurity incidents and breaches that have taken
place over decades. Let us now discuss the major shifts that have changed
cybersecurity over decades:

Personal information or Personally Identifiable Information (PII)


Has become ubiquitous as a result of social media (which
occupies a major portion of our time).
Has become the most precious product in the current marketplace.
Is available in vast amounts and has reshaped the way hacktivists
operate (in the earlier days, attackers could have attacked network
infrastructure and developed ways to get around firewalls,
however, it took time for them to realize that personal information
could be an obvious target).
Is freely available on the web, hence attackers are increasingly
using phishing strategies for data skimming.

There is a dire need to take reasonable measures to protect personal data,


like never before.
People realized, for the first time, how easily personal data can be shared
and spread without their consent.

WannaCry ransomware (2017 event): The first ransomware, the


biggest and most wide-ranging:
Infected over two hundred thousand computers across 150+
countries, holding computers hostage and demanding Bitcoin
payments to return them to their owners.
WannaCry showcased a new age of data risk that urged users to sit up and
take notice like never before. The attack proved that data breaches were no
longer restricted to large enterprises (businesses); they could also invade
and damage personal computers.

Multi-factor authentication
Due to numerous massive data breaches (high-profile
cyberattacks), the use of MFA has been widespread.

MFA protects data and it is now almost mandatory (become the default) for
all websites to implement the same.
We have covered a brief history of cybersecurity (1940-Present). We then
looked at some notable cybersecurity events from 2010 to date) and then
delved into what we believe were the major shifts that had a remarkable
impact on cybersecurity.
In summary, these times have witnessed a sea change concerning
technological advancements and their impact on cybersecurity.
Cybercriminals have continuously been bestowed with new opportunities to
exploit due to the increase in connectedness and the ongoing digitization in
most aspects of life. It is also evident that cyberattacks and Cybersecurity
measures have become sophisticated with time passing.
Here is another dimension to take cognizance of. With the world as
connected as it is, let us not forget that Cybersecurity is not only about
protecting computer systems but also about protecting people. People have
weaknesses, and, like computers, these vulnerabilities can be taken
advantage of. Emotional manipulation and social engineering are being
used by hackers to gain access to otherwise secure systems.
Discussion would be incomplete, if we do not mention that in
Cybersecurity, we must learn from our past mistakes, and apply the lessons
learned to prevent attacks in the future. This is where professionals such as
security researchers and ethical hackers come into the picture. The
collective effort of Cybersecurity professionals should be to discover and
fix vulnerabilities before they are exploited and help make us and our
computers safe.
Cybersecurity threats evolution
Cybersecurity threats have been advancing from time immemorial. This
section presents a unique perspective of how the threat landscape of cyber
threats has evolved periodically (say shifted constantly) over time. Up until
this point, we have seen the following five Life Stages (LS) of
cybersecurity threats:

LS 1: Virus
Refer to section Decade of Commercial Antivirus – The
1980s of this chapter, where we discussed how virus
attacks against standalone computers prompted the
creation of the first antivirus.
LS 2: Network
Refer to section Connected World – The 1990s, the
decade that gave us The Internet (growth and
development). When cyberattacks started coming in
through the Internet, a Firewall was designed to detect
and stop them.
LS 3: Applications
Due to the widespread exploitation of vulnerabilities
inside application software, security measures such as
the implementation of Intrusion Prevention Systems
(IPS) gathered pace.
LS 4: Payload
From the 1980s when the world saw the early
emergence of malware specimens till the advent of
Petya and NotPetya (the encrypting malware), malware
became more targeted. It was able to circumvent
signature-based defenses, and solutions such as Anti-bot
and Sandboxing became necessary to detect these
threats. Therefore, we refer to this advanced threat as
payload malware.
LS 5: Multi-vector
Most recent cyberattacks are large-scale and employ
multi-vector techniques. The growth in sophistication of
cyberattack techniques has rendered the previous forms
of Cybersecurity measures (solutions) less effective
over time. Cybersecurity solutions for LS 5 threats are
required to protect against the present cyber threat
landscape adequately. All we need now is advanced
threat protection solutions.

Building blocks of cybersecurity


After all that we have discussed and learned in this chapter, cybersecurity
revolves around data. The focal facets to be considered are - how is data
stored, who has access, what type of access, and how data is transferred.
The most integral aspect is to ensure that the integrity of data is maintained
at all times, failing which will render data useless. Protecting the integrity
of data should on the forefront of any comprehensive security system.
Equally crucial is to harden the security ecosystem with proper access
controls. Access to sensitive information should be controlled by applying
appropriate access permissions (preventing unauthorized access). Access
controls also play an important role in monitoring and logging an entire IT
ecosystem.
Identity and Access Management (IAM) needs to be mentioned which
allows enterprises (businesses) to organize teams in a security-centric way.
Due to changes in roles, personnel, and business needs and goals over time,
data access is a moving target. Data access permissions should reflect these
dynamic changes.
Finally, availability, of not just data but of services and tools, is critical to
the day-to-day functioning of an enterprise. All network-based threats,
DDoS, and Malware attacks need to be remediated, or else they have the
potential to shut down entire IT systems.

Traditional cybersecurity measures


As the threat landscape has evolved, the cybersecurity measures have
advanced as well, however as point solutions (i.e., specific to threats,
targeting single elements of the security landscape). It would be reasonable
to state that cybersecurity measures have been siloed. This method of
securing IT systems involves pinpointing a problem area, such as the need
for a Firewall, and using a single solution to fill the one defense gap. This
creates a silo.
We mentioned IAM solutions above that cover access controls
(permissions), authorization, and authentication. However, IAM is just a
one-part solution in any comprehensive security ecosystem. Similarly,
network security tools, like perimeter defense solutions, traffic flow
monitoring, intrusion detection, and DDoS protection, cover another section
of an enterprise’ (business’) security needs. Likewise, encryption, web
application security, threat intelligence, data loss prevention solutions, etc.
are some other types of security solutions that are part of a holistic
cybersecurity ecosystem.
Considering that the ecosystem has become too complex, cybersecurity
silos are not effective. The advent of a hybrid work environment (home
workers, on-premises resources) and technologies such as the Cloud (edge
devices) have added to a complex web of security risks. Siloed treatment,
therefore, leads to a lack of visibility and control.
A major issue with using a siloed approach is that attackers can avoid
detection. For example, an attacker might use a trusted platform, like a
penetration tool to install malware or create a communication channel
between a trusted server and a malicious server. Next, the attacker
exfiltrates data using the communication channel to the malicious server.
Assuming the trusted server regularly sends data to external servers, this
kind of data transfer will not seem unusual. What is missed is the fact that
penetration testing may set up communication channels for the duration of a
test but not keep the channel open for long periods. By using siloed security
information systems, we are not able to see connections between events that
would look suspicious and prompt further investigation.
Response times in siloed security systems are too slow. These systems are
difficult to scale and do not talk to each other, leading to a lack of precision
required for remediation and detection. The possibility of human error is
also much higher than when using an integrated security solution. Human
error can also result from alert fatigue, which in turn can lead to serious
lapses in threat preparedness.
Conclusion
It is evident that with the evolution of technology, the domain of
cybersecurity has evolved and so have criminals and bad actors whose
endeavor is to exploit weaknesses in the system to prove a point for
personal gains. It is also important to note that the threat actors have
continued to thrive against the backdrop of macroeconomic and geopolitical
uncertainties. They have incessantly used all the tools and intelligent
resources available at their disposal to make their way past corporate
defenses.
We have covered a brief history of cybersecurity (1940-Present). We then
looked at some notable cybersecurity events from 2010 to date) and then
delved into what we believe were the major shifts that had a remarkable
impact on cybersecurity. In summary, these times have witnessed a sea
change concerning technological advancements and their impact on
cybersecurity. Cybercriminals have continuously been bestowed with new
opportunities to exploit due to the increase in connectedness and the
ongoing digitization in most aspects of life. It is also evident that
cyberattacks and cybersecurity measures have become sophisticated with
time passing.
Discussion would be incomplete, if we do not mention that in
Cybersecurity, we must learn from our past mistakes, and apply the lessons
learned to prevent attacks in the future. This is where professionals such as
security researchers and ethical hackers come into the picture. The
collective effort of Cybersecurity professionals should be to discover and
fix vulnerabilities before they are exploited and help make us and our
computers safe.
The need of the hour is for each siloed system to be a part of the larger
security picture. Network security tools, IAM solutions, Endpoint security,
etc. should all be visible and manageable from a centralized place. Since,
modern threats have ways of countering or avoiding single-point security
controls, a comprehensive view of the enterprises’ (business’) security
environment is a must to detect today’s attacks.
In the next chapter, we will understand today’s cybersecurity challenges.

Points to remember
Cyber-attacks take pace daily and are evolving in sophistication
constantly
Milestone cybersecurity events have played a major role in what
cybersecurity is today
Everyone is susceptible to cyber-attacks
Cyber-attacks perpetrate from insiders as well (in addition from
outside)
It is imperative to stay alert and keep up with important industry trends
to keep attacks at bay
Cyber-attacks will continue to rise in future with their impact
becoming more consequence

Key terms

Phone phreaks: Various practices used by Phreaks or those who have


interest in the functioning of phones and tampering with their
protocols.
Advanced Research Projects Agency Network: Referred to as Early
Internet (network that was built before Internet was created).
Creeper: World’s first computer worm (a code that could be
transported between machines).
Reaper: World’s first anti-virus.
Trojan horse: Type of malware that downloads onto a computer
disguised as a legitimate program.
The orange book: Trusted computer system evaluation criteria
published by the US Department of Defense.
Denial of service: Type of cyber-attack wherein a malicious actor
renders a device unavailable by interrupting the normal functioning of
the device.
Polymorphic virus: Code that alters itself while spreading through
computing systems.
Secure sockets layer: It is a protocol to keep people secure online data
and securely use Internet transactions.
Hyper text transfer secure: Is the secure version of HTTP (protocol
that enables transfer of data between web browser and website.
HTTPS encrypts the data transfer.
Internet chat relay: This is a chat system based on text for instant
messaging.
Zero day attack: Attack exploiting holes in security for new
applications and software (exploiting the flaw before they are
addressed by the developers)
General data protection regulation: A compliance regulation that
provides citizens of the EU greater control over their personal data.
Insider threat: Cybersecurity threats that begin with authorized users
viz., employees/ contractors/business partners.
Supply chain attack: Cyberattack against enterprises (businesses)
suppliers.
Multi factor authentication: Authentication method, a core aspect of
a strong identity and access management framework. It mandates a
user to provide two or more verification factors to gain access to a
computing resource.
Distributed denial of service: Type of DoS attack that involves
multiple connected online devices.
California consumer privacy act: A data privacy law in California,
US to protect the data and privacy rights of residents.
Personally identifiable information: Any data that could potentially
identify a specific individual.
Identity and access management: A framework that aims to control
user access to critical information within enterprises (businesses).
Artificial intelligence: The ability of a computer or robot (computer-
operated) to perform tasks commonly associated with human beings.
Machine learning: The field of AI that focuses on using data and
algorithms to behave the way humans learn.
References

1. https://cybermagazine.com/cyber-security/history-cybersecurity
2. https://www.digitalguardian.com/blog/biggest-moments-
cybersecurity-history-past-10-years

Join our book’s Discord space


Join the book's Discord Workspace for Latest updates, Offers, Tech
happenings around the world, New Release and Sessions with the Authors:
https://discord.bpbonline.com
CHAPTER 2
Cybersecurity: Understanding Today’s
Security Challenges

Introduction
Through Chapter 1, Cybersecurity: A Dynamic Changing Paradigm, we
realize that the modern cybersecurity threat is complex, diverse, and
dynamic. Since time immemorial, we have seen new threats emerge every
year. While attackers continuously seek new ways to bypass defenses,
cybersecurity professionals are finding newer ways to secure sensitive data.
It would be fair to state that people on both sides of the equation (attackers
and defenders) are smart and determined.
One of the most significant developments in the cyber world, especially in
the last decade, is the expansion of the attack surface. Each end-point
(employees working from the office or home) or perimeter device (edge
devices, hybrid cloud systems) in an enterprise (business) is a potential
entrance (unauthorized) into its internal systems. What exacerbates the
issue is their growing numbers.
Let us now concentrate on the systems in an enterprise (business) that are
widely distributed in nature and on how their disparateness impacts
Cybersecurity. The modern-day infrastructure is not only disparate but also
increasing in breadth, which presents new and unique challenges to
cybersecurity operations.

Structure
In this chapter, we will cover the following topics:

Distributed systems
Security challenges of distributed systems
Digital age: Cybersecurity threats
Digital age: Cybersecurity key issues

Objectives
This chapter looks at how disparately distributed systems operate in today’s
enterprises (businesses) and their impact on cybersecurity. It will also
explore varied cybersecurity challenges (threats, attack vectors, and issues).

Distributed systems
Since there is never a one-size-fits-all solution, all large enterprises
(businesses) have highly distributed IT systems and face the consequences.
Each enterprise (business) has definite organizational and operational needs
that must be met, unique ecosystems that must be hardened, and limited
resources available to commit to Security Operations (SecOps).
Consider this common scenario through which enterprises (businesses)
distribute their systems: multiple campuses | Branch offices | Multiple
public clouds | Edge computing | IoT devices, etc. Many larger enterprises
(businesses) will have this distribution or may combine some of these
aspects. Now, think about the administrative complexities of managing
these resources.
For instance, say your business operations are spread across several offices.
Each office’s devices, network, and on-premises computing resources will
need to be secured. The enterprise (business) would likely be using cloud
technologies in day-to-day operations given its advantages in doing so.
Additionally, for each office, communications between public clouds and
on-premises devices/applications would need to be secured. Consequently,
the volume of data will increase, leading to an increase in log data to sift
through. There will also be a dire need for the right-set visibility of data
across the enterprise (business). This is how inherently the ecosystem of
distributed systems works.

Relatively easy to permeate


In distributed systems, various sub-systems (e.g., network
communication/user or device authentication) need to communicate with
one another. Both networks and authentication events need to be monitored.
Moreover, edge computing, remote offices, and other external devices need
access to company networks and systems. This means opening the door for
legitimate communication.
Authentication presents a similar problem. Each user or device attempting
to access the network has to be treated as a potential security incident.
Given the rise in phishing attacks, user devices are particularly vulnerable
to becoming carriers of malicious code. Legitimate users can be hijacked by
attacker code, not only from phishing threats but also from compromised
web applications that inject malicious code into user devices.
Public cloud is one of the biggest security concerns nowadays. Public cloud
adoption leads to an increase in configuration and management overhead,
new authentication challenges, and increased network monitoring
requirements.

Security challenges of distributed systems


The main problems with distributed ecosystems that make it difficult to be
hardened at all times are that they are relatively easy to permeate, lack
visibility, and are difficult to manage, given their dispersed nature.
Let us take these problems one at a time.

Lack visibility
The ecosystem of any distributed system comprises many applications,
servers, and services. This makes the ecosystem difficult to monitor. On one
hand, there are many individual elements to monitor, and on the other hand,
there is log overload and complexity of communication that make it
difficult to see what is transpiring.
Traffic and activity monitoring thus becomes the main pain point as they
maneuver through the web of interconnected systems to find a threat’s
source. Though modern security solutions such as Security Information
and Event Management (SIEM) make it easy to monitor data, analyzing
this data is another story. The problem is simply the volume of data
generated with SIEM and other solutions. Owing to the overwhelming
amount of monitoring data, there is every chance of potential security
events (probably breaches) getting lost and unusual usage patterns getting
overlooked. Security professionals also suffer from burnout trying to sort
through the deluge of information.
The real problem is the visibility of events, not the number of alerts. Data
integration is a huge challenge in siloed security solutions (since a large
volume of logs and monitoring information comes from different
tools/separated regions of the ecosystem), leading to delays in properly
integrating that data.
Cybersecurity professionals do not have the luxury of time if a threat event
materializes. Any enterprise’ (business’s) effectiveness in responding to an
attack (its readiness) is assessed through a metric referred to as Mean Time
To Detect (MTTD). MTTD is the average time taken to detect a security
incident or failure from the time it takes place in a system. The longer a bad
actor has access to internal systems, the more damage they can do and the
more money and time it will cost an enterprise (business).
Another important metric in this context is Mean Time To Remediate
(MTTR). MTTR is the amount of time taken by an enterprise (business) to
neutralize an identified threat or failure within their network environment.
Threat remediation is the process used by enterprises (businesses) to
identify and resolve threats to their network environment. A threat is a
malicious intrusion/infiltration into a system to steal information (that
negatively affects operations and/or damages hardware or software).
MTTD and MTTR are vital metrics to consider when assessing the
effectiveness of any enterprise’s (business’s) security systems. With
maturity in the cybersecurity posture, these metrics should follow a
downward trend as time progresses.
Visibility is not a matter of knowing that an attack is occurring, what
matters is whether we have a proper understanding of the threat and
whether we have the required context. If, due to a vulnerable port in the
infrastructure, a network security alert is noticed, merely closing that port
might not be the whole solution. It is important to understand why that
specific network was compromised. Comprehensive visibility is required
for the cybersecurity staff to understand why the network was compromised
(was the device infected with malicious code?). In summary, modern IT
ecosystems are large and interconnected, so comprehensive visibility is the
key.

Difficult to manage
Distributed systems require well-defined policies to function properly
(policy consistency is difficult to maintain across siloed systems). In
security Silos, there are different tools/ solutions/resources with different
policy control requirements and needs. The policy controls may need to
change over time depending on operational requirements and the changing
threat landscape. In these scenarios, it is a must to stay agile and adapt to
new security threats. Essentially, every day sees new threats emerging and
changing attack patterns, prompting security teams to change strategies to
meet these new attacks.
Inconsistent application of policies is likely to produce redundant/false
alerts, leading to adverse impacts. Also, in siloed security systems, any
fixes (not well thought through) to complex problems are likely to create
new vulnerabilities.

Digital age: Cybersecurity threats


Actions performed by individuals with wrongful intentions to steal data,
cause harm, or disrupt computing systems are treated as cybersecurity
threats. Threat categories cover malware (ransomware), social engineering,
man-in-the-middle (MitM) attacks, denial of service (DoS), injection
attacks, etc. These threats can come from various sources, such as hostile
nation-states, terrorist groups, individual hackers, and trusted
employees/contractors, who take advantage of their privileges to perform
malicious acts.

Sources of cyber threats


Listed are some of the several common sources of cyber threats against
enterprises (businesses):

Nation states: These are hostile countries working to interfere with


communications (disrupt other nations) and gain secrets. They are
politically/economically motivated to launch cyber-attacks against
companies and institutions.
Terrorist organizations: These are terrorists that perpetrate cyber-
attacks to harm/destroy critical infrastructure, threaten national
security and disrupt economies.
Criminal groups: These are a group of hackers who are well
organized to break into computer systems for economic benefit.
Phishing, malware, spamming, and spyware are some techniques used
for extortion, theft of private information, and online scams.
Hackers: These people are usually motivated by personal gain,
financial gain, political activity, or even revenge and use various attack
techniques.
Malicious insiders: This is an employee with legitimate access to
company resources who abuse their privileges to steal information or
damage computing systems for economic or personal gain. Insiders
can be employees, contractors, suppliers, or partners of the target
organization. They can also, in some cases, be outsiders who may have
compromised a privileged account and are impersonating its owner.

Digital age: Cybersecurity attacks


So far with what has been discussed in this chapter, technology is taking
over the business. This is because technology has a huge influence on
business decision-making and reducing manual work.
Millions of people’s personal information can be exposed as a result of a
single security breach. These circumstances can lead to adverse business
financial impact and loss of client confidence. Therefore, it is critical to
have cybersecurity to safeguard both persons and enterprises (businesses)
from spammers and online criminals.
With this backdrop, let us look into some of the challenges facing
enterprises (businesses) in the digital age. The following figure lists various
attacks, which are subsequently discussed further:
Figure 2.1: Digital age cybersecurity attacks

Malware attacks
Malware, also known as malicious software, is the most common type of
cyberattack and covers viruses, worms, trojans, spyware, and ransomware.
Malware perpetrates a system via a link on an untrusted website/email/an
unwanted software download. It has a typical way of operation, firstly it
gets deployed on the target system, then starts collecting sensitive data,
manipulates the system, and may destroy data/or shut down the system
altogether.
Here are some of the main types of malware attacks:

Viruses: This is a piece of code that injects itself into an application


and on application run, executes the code.
Worms: This is a piece of malware that exploits software
vulnerabilities and gains access to computer systems through
backdoors. The worms (malware) are behind most of the Distributed
Denial of Service (DDoS) attacks.
Trojans: This is software (malicious code) that looks to be a harmless
program and hides itself in apps/games/email attachments. When a
user (gullible) downloads the trojan, it gains control of the user’s
device.
Ransomware: It is a type of malware in which the data on a victim's
computer is locked, and payment (monetary ransom) is demanded
before the data is unlocked (using a decryption key). There is no
guarantee that paying the ransom will lead to the restoration of full
access/functionality.
Cryptojacking: This is a type of attack where software is deployed on
a user’s computer, thereby using its computing resources to generate
cryptocurrency (without the user's knowledge).
Spyware: This is an attack where a threat actor gains access to a user’s
(gullible) sensitive information (via desktop browsers/applications and
mobile phones).
Adware: Adware is related to Spyware but it does not install software
on users’ computers and is not used generally for malicious purposes.
However, it can be used without the users’ consent and tracks users’
browsing activity to determine behavior patterns and interests thereby
allowing advertisers to send the user targeted advertising.
Fileless malware: This is an innovative type of malware wherein no
software is installed on users’ computers and is stealthy in nature.
Malicious functions are enabled through editing native files such as
WMI and PowerShell. These attacks are difficult to detect, and the
computer system believes the compromised file is legitimate. These
attacks cannot be detected by traditional antiviruses.

Ransomware attacks
Ransomware is a type of malware in which the data on a victim's computer
is locked, and payment (monetary ransom) is demanded before the data is
unlocked. After successful payment, access rights are returned to the victim.
These attacks are perpetrated by cybercriminals or nation-state-sponsored
groups.
The earliest ransomware attacks simply demanded a ransom in exchange
for the encryption key needed to regain access to the affected data or use of
the infected device. By making regular or continuous data backups, an
organization could limit costs from these ransomware attacks and often
avoid paying the ransom demand.
However, in recent years, ransomware attacks have evolved to include
double-extortion and triple-extortion attacks that raise the stakes
considerably—even for victims who rigorously maintain data backups or
pay the initial ransom demand. Double-extortion attacks add the threat of
stealing the victim’s data and leaking it online; on top of that, triple-
extortion attacks threaten to use the stolen data to attack the victim’s
customers or business partners.
These attacks result in a crisis-level operational impact on critical
infrastructure and commercial enterprises (businesses), while criminals
threaten to publicly release or destroy data if prompt payment is not made.
In the past decade, ransomware attacks have evolved from a consumer-level
nuisance of fake antivirus products to sophisticated malware with advanced
encryption capabilities that now primarily target public and private-sector
enterprises (businesses).
Also, ransomware is now one of cybercrime’s most profitable and popular
business models. It has proven effective and profitable for enterprises.
Enterprises (businesses) need a powerful recovery strategy against
ransomware attacks. This involves proper planning to recover corporate and
customer data and applications.
Take a look at this, cybercriminals do not necessarily need to develop their
own ransomware. Some ransomware developers share their malware code
with cybercriminals via Ransomware-as-a-Service (RaaS) arrangements.
The cybercriminal, or affiliate, uses the code to carry out an attack, and then
splits the ransom payment with the developer. It is a mutually beneficial
relationship: affiliates can profit from extortion without having to develop
their own malware, and developers can increase their profits without
launching additional cyberattacks.
One of the best countermeasures against ransomware attacks is Disaster
Recovery as a Service (DRaaS) solution. DRaaS is a cloud computing
service model that allows an organization to back up its data and IT
infrastructure in a third-party cloud computing environment and provide all
the DR orchestration, all through a SaaS solution, to regain access and
functionality to IT infrastructure after a disaster (e.g., ransomware Attack).
With DRaaS solutions, enterprises (businesses) can automatically back up
files, easily identify which backup is clean, and launch a fail-over with the
press of a button when malicious attacks corrupt our data.
See how things evolve. Many had touted 2020 as the Year of Ransomware.
Then came 2021, 2022, and 2023 (and the trend continues).
Listed below are two reports about ransomware attacks:

Verizon's report found that ransomware was involved in 24% of all


breaches.
Sophos' The State of Ransomware 2023 found that 66% of
organizations experienced a ransomware attack in the past year.

Ransomware threats cannot be ignored, especially as attackers are evolving


their tactics.

Cryptojacking attacks
Cryptocurrency has become popular today. Cybercriminals hijack home or
work computers to mine for cryptocurrency (like Bitcoin). Mining requires
lots of computer processing power. Hence, hackers can make money by
stealthily piggybacking on someone’s computing resources. For enterprises
(businesses), cryptojacked systems can cause serious performance issues
and costly downtime.

Supply chain attacks


Supply chain attacks rose to prominence in late 2020, grew through 2021-
23, and are likely to continue to be a major threat in 2024. SolarWinds
(refer to section The Solar Winds Attack from Chapter 1, Cybersecurity: A
Dynamic Changing Paradigm) hack reported in December 2020 led this
trend. It involved nation-state actors exploiting an IT performance
monitoring system and gaining access to more than 30,000 SolarWinds
customers and partners. Threat actors compromised SolarWinds’
development environment and inserted backdoor code into its Orion
network monitoring product. The discovery of the Sunburst malware kicked
off an extended investigation that uncovered the details of the SolarWinds
hack, multiple malware variants, and an attack campaign that impacted
various public and private sector organizations.
Another high-visibility supply chain exploit in 2021 was the Kaseya Attack,
which leveraged the relationships between Managed Service Providers
(MSPs) and customers to distribute ransomware using MSPs’ remote
monitoring and management software.
In this light, recount the Log4j zero-day vulnerability exploitation (Log4j is
a widely-used Apache logging library). The zero-day vulnerability in the
Java-based Apache Log4j library allowed an attacker who could control the
contents of log messages or their parameters to achieve remote code
execution. This flaw was widely exploited with about 40,000 attempted
attacks detected within two hours of it becoming public and over 830,000
attempts within the first three days. While enterprises (businesses) could
quickly update the library version they used, the libraries used by their
suppliers and partners - and in turn their suppliers and partners, and their
suppliers and partners, and so on - needed to be updated to avoid being
vulnerable to attack.
The high-profile supply chain attacks have demonstrated that it is a viable
and potentially profitable attack vector for cyber threat actors. Going into
the future, cyber threat actors are likely to expand their use of supply chain
attacks to amplify the reach and impact of their attacks. Organizations must
be mindful of the third-party vendors and suppliers they work with. Trust is
an inherent value here, but organizations should also do their due diligence
in vetting third parties.

Social engineering attacks


Not only are the hackers continually evolving their tactics in the use of
technology, but they are also now using weaknesses in psychology to their
advantage. Social engineers are hackers who exploit weaknesses in human
psychology. Hackers use phone calls and social media to trick people into
offering them access to sensitive information.
Let us examine Phishing, the most common type of social engineering
attack.

Phishing attacks
A Phishing attack is a type of social engineering attack that targets users’
login details and credit card information (this user information benefits the
malicious actors). These attacks involve malicious actors to trick employees
into revealing sensitive information in many forms. Those forms include
email phishing and the more sophisticated and targeted spear phishing,
Business Email Compromise (BEC), whaling, and vishing attacks.
Here are some of the main types of phishing attacks:

Spear-phishing is a type of phishing attack that targets specific


individuals or organizations, typically through malicious emails.
BEC is a form of phishing attack where a criminal attempt to trick a
senior executive (or budget holder) into transferring funds or revealing
sensitive information.
Whaling is a highly targeted phishing attack aimed at senior
executives- masquerading as a legitimate email.
Vishing is short for voice phishing, which involves defrauding people
over the phone, enticing them to divulge sensitive information.

Phishing attacks are a never-ending challenge faced by organizations of all


shapes and sizes, and no company nor employee is immune to attack.
According to Verizon’s 2023 Report, 18% of all breaches involved were
phishing.
Let us simplify this for the readers. If you are a Gmail (Google email
service) user, you might come across a spam folder consisting of emails the
platform recognizes as a threat to your data security. These spam emails
consist of thousands of phishing attacks that your mailing partner
recognizes and warns you about the potential cyber threat that it carries.
Yet, some communications still make it to your inbox, where you might fall
into a trap. Officially, Google released a statement of how it blocks over
100 million phishing emails every day. It further emphasized how most of
the communications were trying to impersonate government officials,
authorities, agencies, or websites to sound more reliable to mail recipients.
The following are some notable phishing attacks:

Facebook and Google were scammed out of more than $100 million
after attackers impersonated a legitimate partner of the businesses
between 2013 and 2015. The phishing scams involved contracts and
invoices for funds due.
Sony Pictures was hacked in 2014 after company executives received
phishing emails from a group named Guardians of Peace. The
attackers reportedly stole more than 100 TB of data.
Austrian aircraft supplier FACC was defrauded of $54 million in 2016
after an employee was phished by an attacker, purporting to be the
company CEO, who requested a wire transfer to a bank account
controlled by the attackers.

Generative artificial intelligence phishing attacks


The year 2023 witnessed the advent of Generative AI (GenAI) platforms,
such as ChatGPT. Along with this came a string of security challenges,
especially when it comes to phishing.
GenAI can improve grammar and spelling to help attackers craft more
convincing social engineering and phishing scams. However, it can also
gather information about people and companies from social media and
other websites to conduct targeted spear phishing and BEC campaigns.
A major AI phishing concern is deepfakes. This type of AI creates fake yet
convincing audio, image, and video content to fool people into believing
their legitimacy. Deepfakes can lead to misinformation campaigns,
blackmail, reputational damage, election interference, fraud, and more.

Internet of Things attacks


Internet of Things (IoT) is a system of interrelated physical devices which
can be accessible through the Internet. The connected physical devices have
a unique identifier (UID) and have the ability to transfer data over a
network without any requirements of human-to-human or human-to-
computer interaction. The firmware and software that is running on IoT
devices make consumers and businesses highly susceptible to cyber-attacks.
Let us view some statistics. In 2024, the number of connected devices is
predicted to increase to more than 14.4 billion. IoT Analytics 5 (a leading
global provider of market insights and strategic business intelligence for
IoT, AI, Cloud, Edge, and Industry 4.0) claims that by 2025, there will be
more than 27 billion gadgets online simultaneously.
The IoT sector is the primary target for hackers accessing users’ sensitive
data. To access your device containing your sensitive information, hackers
use devices (IoT gadgets) that surround you, such as wearable
smartwatches, baby monitors, smart fridges, smart lights, smart watches,
webcams, household appliances, medical devices, automobiles, and even
home security systems.
With the expansion of IoT gadgets, it is pivotal to guarantee their security.
These gadgets are vulnerable because they often have default setups, lack
firmware refreshes, and suffer from inadequate encryption. To avoid
unauthorized access and potential breaches, IoT devices must have robust
authentication, encryption, and monitoring mechanisms.

Smart medical devices and Electronic Medical Records vulnerabilities


The healthcare industry realizes the benefits of developments in smart
medical devices. With this, patient medical records have now moved online.
This has led to concerns about cybersecurity threats, privacy, and safety.
As more devices are connected to hospitals and medical facilities, hackers
exploit vulnerabilities in their security defenses. This is rendering patient
data and information (sensitive information) increasingly vulnerable. The
possibility of remote compromise of a device directly connected to a patient
always exists. An attacker could theoretically increase or decrease dosages,
send electrical signals to a patient, or disable vital sign monitoring.

Privacy concerns of connected cars


As technology evolves, the connected car is becoming more and more
ubiquitous. Connected car provides access to various infotainment services,
such as music, radio, movies, news updates, and text/WhatsApp
notifications. These cars support navigation systems through third-party
apps, ensuring real-time traffic updates and route guidance for efficient
navigation to destinations.
For hackers, this evolution in automobile manufacturing and design means
yet another opportunity to exploit vulnerabilities in insecure systems, steal
sensitive data, and harm drivers. In addition to safety concerns, connected
cars pose serious privacy concerns.

Cloud attacks
Cloud computing is an innovative modern-day technology that
revolutionized the physical world of data storage. Cloud computing has led
to improvements in IT efficiency, and provided flexibility and scalability to
enterprises (businesses). Due to these advantages, enterprises (businesses)
from large to Micro, Small, and Medium Enterprises (MSMEs) utilize
cloud services for storing their data (that includes Corporate and user-
sensitive information). On the one hand, its adoption promises a reduction
in cost and increased efficiency; on the other, it opens possibilities for data
security breaches.
In cloud computing, enterprises (businesses) must understand the Cloud
shared responsibility model. The model is a working framework followed
by Cloud Service Providers (CSP) that details the responsibility over an
entire cloud environment, from infrastructure to hardware, data, identities,
workloads, network, settings, and more. Responsibility is divided between
the CSP and the customers. Many organizations cannot delineate where
CSP responsibilities end and their responsibilities begin, opening them to
numerous vulnerabilities.
The exploitation surface for cyber criminals has widened due to using cloud
computing services, and enterprises’ (businesses’) need to address this. The
main reasons for compromised data security/that present an opportunity for
cyber criminals to exploit are the lack of encryption, multi-factor
authentication, improper configuration, insecure APIs, poor access control,
shared tenancy, and supply chain vulnerabilities.

State-sponsored attacks
Hackers are eyeing to make a profit by stealing individual and corporate
data. Now, nation states are using their cyber skills to infiltrate other
governments and perform attacks on their critical infrastructure.
Cybercrime, which has been a major threat to the private sector and for
individuals so far, has now become a menace for the governments and
nations. As we move into the future, state-sponsored attacks are expected to
increase, with attacks particularly on critical infrastructure.

Insider attacks
All cybersecurity challenges in enterprises (businesses) do not come alone
from the outside; they come from within as well. Insider threats are threats
initiated by authorized users (who could be employees, contractors, or
business partners) who advertently or inadvertently misuse their legitimate
access or have their accounts hijacked by cybercriminals.
While it is the external threats that get noticed (become headlines), insider
threats, whether malicious or the result of negligence—can be costly and
dangerous for enterprises (businesses). For example, confidential data may
be leaked by employees (which can prove extremely detrimental as the data
can be used by its competitors. As a result, it can bring significant losses to
the company’s finances and reputation.
Look at these reports that indicate that data breaches due to insider threats
are costlier than data breaches due to external threats:

IBM’s 2023 report: average cost of a data breach ($4.90 million due
to insider threat versus $4.45 million due to external threat).
Verizon’s 2023 report: average compromises due to external threat of
~200 million records versus 1 billion records or more due to insider
threat.

Man-in-the-middle attack
This attack involves intercepting the communication between two
endpoints, such as a user and an application. The attacker can eavesdrop on
the communication, steal sensitive data, and impersonate each party
participating in the communication.
Some examples of MitM attacks include:

Wi-Fi eavesdropping: Here, an attacker sets up a Wi-Fi connection


(seemingly looking authentic) that people at large may connect to. The
fake Wi-Fi allows the attacker to monitor the activity of connected
users and intercept sensitive data (login credentials, etc.).
Email hijacking: Here, a legitimate email address of an organization
(say that of a Bank) is spoofed by the attacker. Users are tricked (by
following instructions they believe have come from the bank, while it
is not true) into giving up sensitive information/transferring money to
the attacker.
Domain name server spoofing: In this, a domain name server
(DNS) is spoofed, which directs a user to a fake website (seemingly
looking authentic). Attack uses this technique to steal the user’s
credentials.
Internet Protocol spoofing: Here, an attacker spoofs an Internet
Protocol (IP) address (which connects users to a specific website) to
pose as a website (which in real is fake).
HTTPS spoofing: The secure version of HTTP is referred to as
HTTPS, but this can also trick the browser into thinking that a
malicious website is safe. The attacker uses HTTPS in the URL to
conceal the malicious nature of the website.

Denial-of-Service attack
This attack overloads the target system with a large volume of traffic,
hindering the ability of the system to function normally. An attack
involving multiple devices is known as a DDoS attack.
DoS attack techniques include:

HTTP flood DDoS: In this attack, HTTP requests (seemingly looking


authentic) are used to overwhelm an application or web server. The
target system is forced to allocate as many resources as possible for
each request.
SYN flood DDoS: This attack uses a transmission control protocol
(TCP) connection. TCP connection requires an synchronize (SYN)
request to be sent to the host with a response from the host with SYN-
ACK acknowledging the request, and finally a response back from the
requester with an ACK. It is possible to mess up with server resources,
by sending SYN requests but not responding with the SYN-ACKs.
This sequence is exploited by attackers.
User datagram protocol flood DDoS: In this attack, user datagram
protocol (UDP) is used to flood a host with UDP packets sent to
random ports. This host then searches for applications on the affected
ports and since the packets are not legitimate, the host responds with
“Destination Unreachable”. This leads to the use of the resources of
the host.
ICMP flood: In this attack, inward and outward-bound bandwidth are
consumed, as the target is overwhelmed with a barrage of ICMP Echo
Request packets. The system is unable to keep pace with the rate of
requests, eventually slowing down.
Network time protocol amplification: This attack uses network time
protocol (NTP) servers that are publicly accessible wherein open NTP
servers are used to send high volumes of UDP traffic to a targeted
server.

Injection attacks
Injection attacks exploit various vulnerabilities to directly insert malicious
input into the code of a web application. Successful attacks may expose
sensitive information, execute a DoS attack, or compromise the entire
system.
Here are some of the main vectors for injection attacks:

Structured Query Language injection: In this attack, an Structured


Query Language (SQL) query is entered into a web form or comment
field to exfiltrate data. Since databases in Web applications are based
on SQL, they are vulnerable to SQL injection. Also, there are also
NoSQL attacks that are targeted against those databases that do not use
a relational data structure.
Code injection: In this attack, malicious code is injected into a
vulnerable web application. The code is executed by the web server as
if it were a legitimate part of the application.
OS command injection: In this attack, command injection
vulnerability can be exploited to input commands for the OS to
execute, which the attack to exfiltrate OS data or take over the system.
Lightweight directory access protocol injection: In this attack,
characters are inputted to alter unsanitized lightweight directory
access protocol (LDAP) queries, making the system vulnerable. Since
LDAP servers may store user accounts and credentials for an
organization, these kinds of attacks can be severe.
XML injection: In this attack, vulnerabilities in legacy XML parsers
are exploited. The attack is perpetrated using specially-made XML
documents (which can be used to traverse paths and remotely execute
code.
Cross-Site Scripting: In this attack, malicious code (using JavaScript)
perpetrates a target browser. The browser executes the code,
redirecting users to a malicious website or stealing users’ session
cookies to hijack the session.

Digital age: Cybersecurity key issues


In addition to the above-mentioned cybersecurity challenges being faced by
enterprises (businesses) in the form of Cyber Attacks, other issues need to
be addressed by the industry at large. A few of them are given below.

Budgets issue
Cybersecurity is seen as a cost center by enterprises (businesses) because it
is difficult to calculate its return on investment (ROI). With time, as its
importance has increased, it has been relatively safe from the perspective of
budget and staff cuts. However, it is not immune from them. Chief
information security officers (CISOs) and security teams can face budget
cuts and reductions in spending (we saw this happen during COVID-19).
High-interest rates environment (as it is now), high inflation, recession
fears, and geopolitical uncertainty are all extraneous factors and will
continue to plague the IT industry (in response to which enterprises
(businesses) resort to budgets and staff cuts. In such circumstances, they
must plan carefully to maintain the security posture of their enterprise
(business) without burning themselves out. It is not easy to get more done
with less.

The skills gap issue


Skills shortage is not new to the world of the cybersecurity industry. We
have been hearing this consistently for many years that the industry needs
more security staff than are available to hire. There is a demand-supply
issue. Budget cuts and layoffs make things worse, as this means the same
amount of work is expected to be completed by fewer staff members.
According to the ISC2 Cybersecurity Workforce Study, the industry needs
an additional four million security professionals to safeguard enterprises
(businesses) against growing threats. Hiring employees with the right skills
and employee retention is a stiff challenge. That is the reality even before
considering potential budget cuts and layoffs.

Artificial intelligence issues


We have discussed that attackers can use AI to drive sophisticated
phishing/other types of attacks. Enterprises (businesses) organizations are
facing many concerns related to the usage of AI:

Users might inadvertently or maliciously input sensitive data, such as


source code, copyrighted material, or confidential business data, into
an AI-powered chatbot, which could lead to data exposure.
Attackers can poison AI models with inaccurate data to fool the
models into believing attack behaviors are not malicious.
Using personally identifiable information or sensitive data to train
large language models can result in data leakage, create data privacy
concerns, or lead to data breaches. Copyright infringement, fraud, and
breach of contract are many legal issues that are a possibility.
AI tools, like any other tool or software, might have vulnerabilities that
attackers can exploit.
Be aware of any nonpermitted, non-company-controlled AI use by
employees, known as shadow AI. Security policies or acceptable use
policies should outline the challenges of shadow AI and prohibit it as
needed. Monitor systems for shadow AI use, and assess and remediate
any risks.

Conclusion
We now realize that cybersecurity risks are everywhere, hence
cybersecurity has become a dominant concern for all enterprises
(businesses), whether large, medium, or small. What is suggested here, is
for enterprises (businesses) to adopt and rigorously follow an industry-
accepted cybersecurity framework (e.g., NIST), conduct regular
assessments, and train their employees consistently to make them security
conscious. This is paramount in creating a secure digital environment.
Enterprises (businesses) should implement modern technologies such as
endpoint detection and response (EDR), adopt strong password
management, and conduct cloud security assessments to counter attacks
such as malware, ransomware, and phishing. Cybersecurity strategies have
to continually evolve and adapt to the changes in our digital frontier as the
threat attack surface expands. Cybersecurity should not be a mere tick mark
in the box. Financial and reputation implications on enterprises (businesses)
due to a cyberattack are immense, hence cybersecurity should be an integral
part of business strategy.
In the next chapter, we shall discuss the emerging cybersecurity trends in
details and understand their common themes. We shall also introduce the
subject of cyber resilience, its building blocks, and its importance for
enterprises (businesses).

Points to remember

Distributed ecosystems have their challenges viz., they are relatively


easy to permeate, lack visibility, and are difficult to manage owing to
their dispersed nature.
Be aware of various types of cybersecurity challenges
(threats/attacks/issues) being faced by your enterprise (businesses)
today.
It is imperative to be up-to-date on the latest cybersecurity
threats/attacks/issues to safeguard your enterprise (business).
The biggest challenge in cybersecurity today is the evolving nature of
cyber threats.
Cybercriminals are continuously formulating new methods and
strategies to exploit vulnerabilities in networks and systems.
The first step to address cybersecurity risks is to create a plan (based
on an industry-accepted framework.

Key terms
Security information and event management: A solution that helps
enterprises (businesses) to detect, analyze, and respond to security
threats before they can have adverse impact on its operations.

Mean time to detect: A measure of how long it takes to discover a


potential security incident.
Mean time to respond: The time it takes to control and remediate a
threat once it has been discovered.
Internet of Things: A network of interconnected devices that connect
and exchange data with other IoT devices and the cloud.
Data privacy: Element of data protection that addresses the proper
storage, access, retention, security, etc. of sensitive data.
Disaster recovery as a service: A cloud computing service that offers
an enterprise (business) an offsite disaster recovery capability in a
third-party cloud.
Internet control message protocol: A protocol that devices within a
network use to communicate problems with data transmission.

References

1. https://www.techtarget.com/searchsecurity/feature/Ranso
mware-trends-statistics-and-facts
2. https://www.cnbc.com/2019/03/27/phishing-email-scam-
stole-100-million-from-facebook-and-google.html
3. https://www.chegg.com/homework-help/questions-and-
answers/case-study-sony-pictures-entertainment-hack-
problem-november-24-2014-hacker-group-called-g-
q45826501
4. https://www.computerweekly.com/news/4500271523/54m
-cyber-fraud-hits-aircraft-supplier-share-price
5. https://www.techtarget.com/iotagenda/opinion/IoT-
trends-to-keep-an-eye-on
6. https://www.ibm.com/topics/insider-threats
7. https://www.nextdlp.com/resources/blog/seven-
takeaways-from-2023-verizon-data-breach-
investigations-report
CHAPTER 3
Emerging Cybersecurity Trends

Introduction
In the previous chapters, we looked at the chronology of the evolution of
cybersecurity, explored noteworthy cybersecurity events and their impact
on this domain, and also examined the building blocks of cybersecurity.
Further, we understood distributed systems, their workings, and challenges.
We then detailed cybersecurity threats, attacks, and key issues and further
discussed how to secure digital boundaries.
In this chapter, we will explore some prominent cybersecurity trends (and
their characteristics) that are present and emerging due to the change in the
threat landscape owing to our ever-increasing reliance on digital technology
for conducting business.

Structure
In this chapter, we will cover the following topics:

Cybersecurity trends
Cyber security future trends
Common themes in trends
Understanding cyber resilience

Objectives
Cybersecurity has become a vital objective for enterprises (businesses) to
protect data from online threats and unauthorized intrusions. Cybersecurity
trends have evolved with technological advances, and data breaches,
ransomware attacks, and cyber hacks have increasingly become
commonplace.

Cybersecurity trends
With the increase in reliance on digital technology for communication,
commerce, and critical infrastructure, the threat landscape has evolved in
complexity and sophistication. In this section, we will explore some of the
prominent (20 nos.) cybersecurity trends and the challenges they pose to
individuals, enterprises (businesses), and governments.
Figure 3.1 depicts the first five (1-5) cybersecurity trends (with trends 6-20
to follow in continuation):

Figure 3.1: Cybersecurity trends 1-5

Trend 1: Tenacious data breaches


Data breaches have taken the world by storm and are unlikely to abate.
They affect both individuals and enterprises (businesses) and present
themselves as a paramount concern. In Chapter 1, Cybersecurity: A
Dynamic Changing Paradigm, we covered regulatory frameworks such as
GDPR and CCPA. The risk of data breaches can be mitigated through
rigorously complying with these regulations and implementing security
measures proactively.

Trend 2: Upsurging ransomware


We have touched upon this threat in the preceding chapters. There possibly
are 120+ varieties of ransomware out there and the lure of monetary
rewards is behind its successful rise (the trend of remote working during the
COVID-19 pandemic added to the woes).
Attackers are expected to perpetrate more targeted ransomware attacks on
vulnerable enterprises (businesses) using sophisticated phishing techniques
and the power of artificial intelligence (AI) and machine learning (ML).
Enterprises (businesses) must remain cautious and implement strategies
proactively to mitigate the risks of a ransomware attack.

Trend 3: Smarter social engineering attacks


Social engineering attacks (refer to Chapter 2, Cybersecurity:
Understanding Today’s Security Challenges, for examples like spear
phishing, BEC, whaling, vishing, etc.) continue to target sensitive user
information through illegal means such as phishing by exploiting human
weaknesses. Hence, these attacks will remain a significant threat to
individuals and enterprises (businesses). A combination of employee
awareness is effective in countering the associated risks through training
and implementing security measures proactively.

Trend 4: Imminent cloud security threats


Due to various benefits ranging from scalability, to cost savings to
efficiency, enterprises (businesses) are embracing cloud services. With
sensitive data stored in the cloud, it has become a key target for attackers. In
Chapter 2, Cybersecurity: Understanding Today’s Security Challenges, we
discussed that the responsibility of data security in and of the cloud is of the
customer and cloud service provider (CSP) respectively. Cloud is
becoming vulnerable due to insecure interfaces, misconfigurations, errors
by end users, etc on account of customer and/or CSP.
As the reliance on the cloud increases, enterprises (businesses) must
implement stringent security measures (protocols) and the rigor of
continuous monitoring to mitigate risks and protect sensitive data stored in
the cloud. Security measures can include regular patching, encryption,
authentication, etc.
Cloud-related vulnerabilities will continue to be one of the biggest trends
impacting the cybersecurity industry.

Trend 5: Threat emanating due to mobile devices


Mobile devices have become rewarding targets for attackers due to their
proliferation. We have witnessed a noteworthy increase in malware and
targeted attacks on mobile banking (since mobile devices are used for
carrying out financial transactions). Mobile device security has become an
integral point in the cybersecurity ecosystem and we are expected to see a
rise in mobile device-specific viruses and malware.
Figure 3.2 depicts the next five (6-10) cybersecurity trends (with trends 11-
20 to follow in continuation):

Figure 3.2: Cybersecurity trends 6-10

Trend 6: Evolving IoT threats


Internet of Things (IoT) and IoT attacks were introduced and discussed in
Chapter 2, Cybersecurity: Understanding Today’s Security Challenges. The
number of IoT devices continues to grow, creating greater opportunities for
perpetrating cybercrime. It is important to note that an IoT device has very
little processing and storage capacity compared to a laptop/smartphone.
Because of this, it is not easy to deploy firewalls/ antivirus/other security
applications to protect them. Consequently, enterprises (businesses) need to
prioritize the security of their IoT devices. This can be ensured through
regularly updating the IoT devices and implementing robust security
measures.
IoT devices are more exposed to external threats in the era of 5G networks.
There is an urgency for 5G manufacturers to reduce the risk of data
breaches and network attacks by developing secure hardware and software
solutions.

Trend 7: Growing power of AI


AI is used to enhance various aspects of cybersecurity. AI enables the
development of mechanized cybersecurity systems (by using ML
algorithms) that can do tasks, such as robust threat detection, natural
processing language (NLP), and face detection. Please note that attackers
can take advantage of using these technologies to perpetrate attacks
(automate attacks, do data poisoning, etc.) to circumvent security protocols.
Though the usage of AI in cybersecurity is a work in progress, it is widely
expected that threat detection and threat hunting abilities will gain
importance in sophistication through the application of AI and ML.

Trend 8: Intensifying cyber warfare by rogue nations


Escalation of tensions between two powerful nation-states fuels cyber-
warfare, which is state-sponsored (this nation-state is then known as a
rogue nation). The target in the situation remains to be the critical
infrastructure of the other nation. Events such as national elections are
increasingly becoming targets of such attacks. We expect an increase in data
breaches and acts of cyber terrorism in 2024.

Trend 9: Improving MFA


Usage of multi-factor authentication (MFA) is becoming a default as it
renders an extra layer of security (improves the security posture) and
mandates users to provide authentication before accessing
accounts/systems. Despite MFA enhancing security, attackers will keep
finding ways to bypass it. Therefore, the world is moving away from using
phone-based MFA towards application-based authenticators (such as
Google Authenticator) and security keys, which are more secure.
Enterprises (businesses) will need to continue this journey of adopting MFA
on priority and shield themselves against cyber threats.

Trend 10: Enhanced automation


Management of the ever-growing volume of data (increasing workloads)
and steadying security processes require more and more automation to be
embraced. Also, timely and efficient responses to cyber threats can be
enabled through automation. The workplaces are already seeing (and
continue to see) the integration of security measures into development
processes, which promises to create more secure software solutions.
Figure 3.3 depicts the next five (11-15) cybersecurity trends (with trends
16-20 to follow in continuation):

Figure 3.3: Cybersecurity trends 11-15


Trend 11: Rise of data privacy
Data privacy is a prominent data security trend on the rise that affects many
aspects of an enterprise (business). The world has witnessed millions of
personally identifiable information (PII) records getting exposed over the
years which has led to the introduction of data protection laws worldwide
viz., General Data Protection Regulation (GDPR), California
Consumer Privacy Act (CCPA), etc. which indicates that data privacy will
continue to be prioritized.
Non-compliance with such regulations will lead to fines (penalties), loss of
customer trust, and bad publicity. Due to these implications, enterprises
(businesses) will, going forward, recruit data privacy officers and
emphasize critical controls such as role-based access control, MFA,
encryption (transit and rest), segmentation of the network, etc. to
continually improve their data privacy posture.

Trend 12: Risks of distributive workforce


COVID-19 had a profound impact on enterprises (businesses) by forcing
them to shift their workforces to performing work remotely, which
happened quite fast. It is expected that a hybrid work environment is likely
to prevail post-pandemic. Therefore, a challenge for enterprises
(businesses) will be to safeguard their workforce, which will work in a
distributed environment.

Trend 13: Rise in insider threats


Insider threats will remain a growing challenge for enterprises (businesses).
Security can be compromised inadvertently or maliciously by employees or
trusted individuals. Enterprises (businesses) are expected to increase their
focus on enhancing employee behavior monitoring and detection
capabilities to proactively identify insider threats. Emphasis will be on
employee awareness and regular training.

Trend 14: Widening IT skills gap


The demand for skilled cybersecurity professionals is expected to further
rise. Consequently, the gap between demand and supply (talent available)
will grow. This gap will present a challenge for enterprises (businesses) to
seek competent experts to manage their Cybersecurity requirements.
Enterprises (businesses) will invest in programs (training) to upskill the
existing workforce and attract new talent. The shortage of cybersecurity
will remain an unrelenting issue.

Trend 15: Rising threats of Deepfakes


Deepfake technology has started to create havoc and is expected to remain a
deep concern. It involves manipulating audio and video to create realistic
but fake content. This can be used to perpetrate social engineering attacks,
spread disinformation through impersonation, etc. Enterprises (businesses)
will invest in deepfake detection tools, employees’ education, and
awareness to protect data integrity and reputation.
Businesses of all sizes, corporate entities, organizations, and even
governments have embraced computerized systems to streamline daily
operations. Consequently, ensuring cybersecurity has emerged as a
paramount objective to protect data from many online threats and
unauthorized intrusions. As technology evolves, so do cybersecurity trends,
with data breaches, ransomware attacks, and hacks becoming increasingly
commonplace. Elevate your expertise by enrolling in security courses led
by industry experts, empowering you with the knowledge and skills needed
for comprehensive data protection.
Figure 3.4 depicts the next five (16-20) cybersecurity trends:

Figure 3.4: Cybersecurity trends 16-20


Trend 16: More supply chain attacks
Attacks targeting supply chains of enterprises (businesses) to compromise
their services and products are expected to be on the upswing. Events like
SolarWinds have demonstrated the devastating impact it can have on
enterprises (businesses) and their customers. This will increasingly focus on
enhancing their visibility into the supply chain and implementing robust
cybersecurity measures to reduce the risk of compromise.

Trend 17: Continued prominence of regulations


Cyberattacks are a threat to national security and can thwart economic
growth, and governments and enterprises (businesses) are aware of this.
Regulations have emerged to address the growing cybersecurity issues and
the impact data breaches have on social and even political fronts.
Compliance with regulations (global data privacy laws) such as GDPR,
CCPA, etc., centered around consumer data privacy, the need for aspects
such as consent management, etc. will become more significant. These
regulations will continue to evolve, urge enterprises (businesses) to adopt
more stringent data protection measures and this subject will be on top of
the agenda of legislators.

Trend 18: Emphasis on incident response and recovery


The development and testing of incident response and recovery plans will
be a focus for enterprises (businesses). The capacity to detect and respond
to cyber incidents and recover from them will be critical to minimizing the
impact of breaches.

Trend 19: Cyber resilience


Cybersecurity and cyber resilience are not the same. Cybersecurity focuses
on preventing attacks, while cyber resiliency is a measure designed to
ensure the continuity of business operations after an enterprise (business)
undergoes a successful data breach. The priority in the future will be to
develop the ability to efficiently recover and salvage data loss and business
downtime.

Trend 20: A board room agenda


With time passing, Cybersecurity has been featured among the top business
risks being faced by enterprises (businesses). Consequently, cybersecurity
remains a challenging area of oversight for corporate leaders and many
boards regularly discuss this subject. It is widely expected that in the future
there will be at least one member on the board with expertise in this
domain.

Cyber security future trends


In this section, we will explore some of the prominent (10 nos.) future
(emerging) cybersecurity trends and the challenges they pose to individuals,
enterprises (businesses), and governments.
Figure 3.5 depicts the first five (1-5) future cybersecurity trends (with
trends 6-10 to follow in continuation):

Figure 3.5: Cybersecurity future trends 1-5

Future trend 1: Zero trust


Enterprises (businesses) have always worked with an assumption that
someone inside the corporate perimeter is to be trusted (implicit trust) and
anyone from outside should be suspected. This has led to data breaches
(since attackers can do lateral movement throughout the network if they can
breach the perimeter).
This is where a zero trust model comes to the rescue. It provides user access
to information based on their identities and roles, regardless of their
location (home/office/anywhere). In this model, authentication and
authorization occur continuously rather than just once at the perimeter.
Unwarranted lateral movement by an attacker can be avoided between
applications, systems, and services through zero trust.
It would not be complete if we do not refer to microsegmentation, a core
principle of zero trust. Microsegmentation enables the bucketing of network
resources in isolated zones. This helps contain threats and prevents lateral
within the network. Also, the application of granular role-based access
(adaptive policies) is possible.
Adoption of what is known as Zero Trust Architecture (ZTA) will
increase as enterprises (businesses) realize the shortcomings of perimeter-
based security models for securing sensitive data. ZTA works on two main
principles: continuous monitoring (verification) and never trust, always
verify.
Figure 3.6 depicts a phased approach (iterative) for adopting a journey to
implement zero trust:

Figure 3.6: Phases to Zero Trust

Future trend 2: Cryptography resistant to quantum computing


The world of technology will witness advancement in quantum computing
technology. With this, the development of cryptography that is resistant to
quantum computing will be vital. Enterprises (businesses) will need to use
cryptographic algorithms that can resist quantum attacks.

Future trend 3: 5G network security


In the years ahead, we will see rollouts of 5G networks in different
countries. With this, the emphasis on the security of 5G networks will
increase. Faster speed and lower latency of these networks will present new
security challenges, which will make IoT devices vulnerable in particular.
Future trend 4: Stronger authentication measures
Stronger authentication measures are likely to evolve in the future viz.,
facial recognition and fingerprint scanning. Behavioral analytics or liveness
detection are being talked about now-a-days. These measures can be used to
prevent spoofing. The usage of such technologies will become more
widespread to protect sensitive information.

Future trend 5: Human-centric security


User-centric security awareness and training programs will be expanded to
reduce the risk of social engineering attacks. Behavioral analytics and user-
focused security tools will help identify unusual user behavior and potential
insider threats.
Figure 3.7 depicts the next five (6-10) future cybersecurity trends:
Figure 3.7: Cybersecurity future trends 6-10

Future trend 6: cyber insurance


We will witness cyber insurance policies getting customized to specific
risks and compliance requirements and the world will see a spurt (growth)
of cyber insurance policies as the need for protection against cyber
incidents from a financial perspective will increasingly be recognized.

Future trend 7: AI-assisted security testing


As the power of usage of AI increases in threat hunting and threat detection
domains, AI will be used to power the conduct of penetration testing and
vulnerability assessments. These activities are likely to become more
sophisticated in detecting weaknesses in systems/applications and enable
addressing security weaknesses on a proactive basis.

Future trend 8: AI-driven cybercrime


We have discussed the usage of AI in cybersecurity. This view is from the
other side of the coin. Cyber attackers will also progressively employ AI
and ML to them more sophisticated (enhanced), bypass security measures,
and make them more challenging to detect.

Future trend 9: Smart cities and critical infrastructure security


The future will see the advent of smart cities and the expansion of
digitization of critical infrastructure. In these scenarios, systems will get
more interconnected, and security measures to protect them will become
critical. Governments/enterprises (businesses) will have to prioritize the
protection of critical infrastructure such as power grids, transportation
systems, etc.

Future trend 10: Ethical hacking and bug bounty programs


Enterprises (businesses) utilize programs like ethical hacking, and bug
bounty programs (crowdsourced) to detect system vulnerabilities (this is
through their security testing capabilities). The trend is gaining traction
among enterprises (businesses) that seek a cost-effective and efficient way
to identify and mitigate security vulnerabilities.
Enterprises (businesses) can leverage the collective expertise of a global
community of ethical hackers to unearth vulnerabilities that traditional
security testing methods may not be able to find. This is giving an impetus
to the bug bounty platforms and aiding in the expansion of crowdsourced
security testing across industries.

Common themes in trends


So far, we have explored the cybersecurity trends (present and future).
Mentioned are some of the common themes amongst these trends:

Sophistication: The increase in the sophistication of cyberattacks is


one of the most distinguished trends in the digital threat landscape.
Attackers are constantly developing newer tactics, techniques, and
strategies to exploit (breach) security systems, disrupt critical services,
or steal sensitive data. There are many factors behind this viz.,
powerful hacking tools at the disposal of attackers, the spread of
cybercrime forums, and the upsurge of hacking groups that are state
nation-state-sponsored.
Attack vectors: We have discussed malware, DDoS attacks,
ransomware, etc. which are the various attack vectors being deployed
for carrying out attacks. Attackers don’t use just one, a combination of
various methods to perpetrate attacks to achieve their objective. For
example, a ransomware attack may commence with a phishing email
and eventually lead to the deployment of malware that encrypts data.
Target: None is spared. Large enterprises (businesses), governments,
and individuals were the early targets of the attackers. The cyber
threats equally loom large on medium and small businesses, healthcare
and manufacturing entities, educational institutions, etc. Attackers
crave to disrupt/steal sensitive information, which is leading to this
diversification of targets and is expected to continue as digitization
further permeates our digital world.

The need of the hour for enterprises (businesses) to combat cybersecurity


attacks is international cooperation on cybersecurity matters and cyber
resilience. Both the subjects are briefly descrived as follows:
International collaboration: There is a growing need for
governments around the world, law enforcement agencies, and
cybersecurity enterprises (businesses) to share threat intelligence to
enable effective tracking down of cyber criminals and their
infrastructure. This can lead to mitigating threats at a global level.
Since digital threats are global in nature, international collaboration
has become paramount.
Cyber-resilient architecture: In our increasingly interconnected
world, the importance of cyber resilience cannot be overstated. Cyber
resilience goes beyond mere cybersecurity; it is about preparing for,
responding to, and recovering from cyber threats in a way that
minimizes damage and ensures business continuity.

Understanding cyber resilience


As discussed earlier in this chapter (refer to Trend 19: Cyber resilience),
cyber resilience is the aptitude of an enterprise (business) to effectively
recover after a cyber-attack. This requires anticipating and preparing for the
attack through detective, preventative and other proactive cybersecurity
measures, and timely and efficient providing incident response. The
prudence of cyber resilience acknowledges that breaches will take place. It
enables enterprises (businesses) to proactively prepare for the same by
emphasizing the importance of salvaging the business impact and ensuring
a swift recovery.

Importance of cyber resilience


A successful cyberattack can be devastating enough and cause huge
financial losses or even shut down an enterprise (business) permanently.
Cyber resilience therefore is vital to identify, assess, manage, mitigate, and
recover from malicious attacks.
Refer to Figure 3.8 which reflects the importance of cyber resilience:
Figure 3.8: Importance of cyber resilience
A comprehensive cyber resiliency strategy helps protect critical systems,
applications, and data and also enables swift recovery and business
continuity. Adaptability, business continuity, and protection of sensitive
summarize the importance of cyber resilience:

Adaptability: Cyber resilience prepares enterprises (businesses) to


adjust to constantly evolving cyber threats (of sophisticated and varied
nature), thereby making sure that the defenses remain effective over
time.
Business continuity: The downsides of a cyberattack are immense. It
can interrupt business operations, inflict reputational damages, and
lead to financial losses for enterprises (businesses). Cyber resilience
ensures that vital business functions/services continue operating even
after a cyber incident.
Protecting sensitive data: Cyber resilience not only focuses on the
prevention of data breaches but also the implementation of security
measures to recover data in a timely and secure manner in case of a
data breach. This aspect is paramount for any enterprise (business).

Building blocks of cyber resilience


Cyber resilience is not monolithic in nature. It consists of multiple
structured steps along with multiple iterations.
Refer to Figure 3.9, which depicts the building blocks of cyber resilience:
Figure 3.9: Building blocks of cyber resilience
Following are the steps an enterprise (business) can take to build and
improve its cyber resilience.

1. Assess risk: The first step is to identify and understand


potential cyber risks. An important step in this regard is to
carry out a comprehensive risk assessment to identify
vulnerabilities and potential attack vectors.
2. Plan: The next step after identification is to develop a
holistic cybersecurity plan that covers strategies for
prevention, detection, response, and recovery.
3. Conduct employee training and awareness: This step is
equally important. Employees need to be regularly trained on
cybersecurity best practices (how to recognize phishing
attempts, use strong passwords, etc.). Cybersecurity
awareness should become second nature within the
enterprise (business) as it directly addresses the employees
who are the first line of defense against cyber threats.
4. Regiment incident response plan: The need for developing
a comprehensive incident response plan needs to be
overemphasized. The response plan should entail detailed
steps (clear communication guidelines, roles, and
responsibilities, how to isolate and mitigate the business
impact) to be undertaken in the event of a cyber incident.
5. Conduct data backup and recovery: Conduct regular
backups of business-critical data and make sure that the data
is secure and accessible in time of need (reliable). This will
expedite recovery in situations involving a ransomware
attack or data breach.
6. Collaborate through information sharing: This aspect has
been discussed in this chapter (refer to the section on
International collaboration). Industry peers and government
agencies globally must collaborate and share best practices
and threat intelligence. The cyber resilience of the digital
ecosystem can be strengthened through this approach.

Conclusion
Cyber resiliency in today’s digital world has become a necessity. It is
therefore important that enterprises (businesses) adopt an approach that is
proactive and adaptive which will enable them to maneuver through the
changing threat landscape. Cyber resilience is an ongoing process that
requires 3Cs (commitment, collaboration, and continuous improvement) to
stay ahead of the curve.
In the next chapter, we will understand Cyber Security Mesh Architecture
(CSMA), examine its need CSMA, and detail the benefits of its adoption.

Points to remember

Enterprises (businesses) globally are facing numerous cybersecurity


threats and attacks that can be expected to increase in volume and
complexity over time
Attackers are continuously seeking newer methods to target and cause
harm to individuals/enterprises (businesses)
Cybersecurity issues continue to evolve due to the above-mentioned
reason
Cybersecurity responses (also detective and preventive measures) need
to evolve in capability to counter the growing threats and to stay ahead
of the game
Be aware of cybersecurity trends and their importance
Understanding these trends can provide key insights into the changing
threat landscape of cybersecurity
Cyber resilience is the capability of an enterprise (business) to enable
business resiliency by thorough preparedness to respond to and recover
from cyber threats
An enterprise (business) that is cyber-resilient can address known and
unknown threats, events, and challenges better than others

Key terms

Cloud service provider: Providers of cloud platforms, infrastructure,


and services.
Deepfakes: A type of AI used to create (by using deep generative
methods) image, audio, and video hoaxes that appear convincing
Zero Trust: A combination of concepts and ideas that fundamentally
shift focus from a location-centric model to a data-centric approach for
modular security controls (between users, systems, data, and assets
that change over time)
Cyber resilience: The ability of an enterprise (business) to limit the
impact of security incidents by deploying and optimizing appropriate
security tools and processes

Join our book’s Discord space


Join the book's Discord Workspace for Latest updates, Offers, Tech
happenings around the world, New Release and Sessions with the Authors:
https://discord.bpbonline.com
CHAPTER 4
The Need for Cybersecurity Mesh
Architecture

Introduction
So far, we have learned how cybersecurity has evolved dynamically over
time, we then concentrated on the cyberattacks and challenges being faced
by enterprises (businesses), and explored prominent cybersecurity trends
emerging due to the change in the threat landscape.
Let us now take a solutions-oriented approach and discuss cyber risk
management. Simplistically, it is about assessing what could potentially go
wrong and then deciding on the best approaches/solutions to
prevent/minimize it. In today's world, both the government and private
sectors must manage any cybersecurity threat.
Let us remember the impact COVID had on enterprises (businesses). It has
pressed enterprises (businesses) to what we refer to as an edgeless digital
infrastructure that changes constantly expanding its attack surface. What
have the enterprises (businesses) done in these situations? They have
employed new technologies across their infrastructure.
We now have distinct environments to take care of viz., work-from-home,
on-premise office infrastructure (traditional), and multi-cloud data centers.
Not only has the complexity of safeguarding this attack surface increased
(through tools) but also the management of these tools has added to the
complexity (which is often done by different teams).
It is reasonable enough to state that the cybersecurity industry is at a tipping
point.

Structure
In this chapter, we will cover the following topics:

The current situation of cybersecurity ecosystem


CSMA explained
Illustrating the layers of CSMA
The need for CSMA
Benefits of CSMA

Objectives
In this chapter, we will introduce CSMA and examine its needs and benefits
for enterprises (businesses). We shall also understand the various challenges
that the cybersecurity architectures of today are facing.

The current situation of cybersecurity ecosystem


The cybersecurity tools/platforms/services of today, which we call the next
generation set, need to work together coherently to manage the complex
environment and dynamically changing (and growing) threat landscape.
According to a Global Data Breach Report 2022 by a leading organization,
an average enterprise (business) runs 45+ security tools (point tools
covering those that monitor applications, network, cloud operations, etc.
Many of these tools were not established with contemporary use cases in
mind. Existing approaches to cybersecurity are not adequate to address
issues of dynamic and complex environments where the sophistication of
threat actors is increasing relentlessly.
With this backdrop, Gartner released a document titled Gartner Top
Strategic Technology Trends for 2022. One of the many trends (e.g.,
Generative AI, Hyper automation, Cloud-native platforms, Data fabric, etc.)
mentioned in this document was Cybersecurity Mesh Architecture.
Cybersecurity Mesh Architecture (CSMA) is a new approach to
cybersecurity coined by Gartner. In this chapter, we will discuss the
circumstances that have brought CSMA platforms to the fore, how these
platforms work, and describe the benefits of using CSMA.
CSMA explained
CSMA is a contemporary scalable (say flexible) approach to cybersecurity
architecture that allows the distributed enterprise (business) to deploy and
extend security controls where needed most. Each tool running in a silo is
not helpful, and CSMA enables various tools to run and interoperate via
common identity, centralized policy management, automation,
orchestration, security analytics, and intelligence.
There is another way of perceiving CSMA. It is an architecture that
provides the base for computers and people to connect securely across
various locations over hybrid and multi-cloud environments, channels, and
varied applications while protecting all digital assets of the enterprise
(business). While doing this, the architecture enables a security posture that
is far more consistent and supports the increased agility of the enterprise
(business).

Note: CSMA is not prescriptive; rather, it is a framework that is interoperable and enables
disparate cybersecurity services to work together.

Illustrating the layers of CSMA


Please refer to the following Figure 4.1, wherein you can see that CSMA
buckets existing cybersecurity tools into various layers, where each layer
can interact with the ones above and below it. This leads to forming a mesh
of functionality that allows distributed systems to communicate with one
another (and is far more beneficial). This is fed into centralized dashboards
and controls that streamline security operations.
Figure 4.1: Cybersecurity Mesh Architecture
Source: Gartner

The need for CSMA


We have seen that historically, enterprises (businesses) have taken an
approach to selecting appropriate (according to the need) solutions such as
Firewalls, Network Access Control (NAC), Intrusion Prevention
System/Intrusion Detection System (IPS/IDS), Security Information
and Event Management (SIEM), Endpoint Detection and Response
(EDR), and more, which have been of great assistance to security
operations teams.
These advanced tools were architected to work independently and have
limited interoperability with each other, which leads to ineffective data and
alert management. In short, these tools act as vital components to the bigger
cybersecurity puzzle.
The following figure lists the various challenges plaguing today’s
cybersecurity architectures:
Figure 4.2: Challenges facing today’s cybersecurity architectures

Challenge 1: Data deluge


All enterprise (business) processes depend on fast and accurate data
collection and processing, and so does cybersecurity. A potential concern
for security teams is data deluge (refer to Figure 4.3, which illustrates a
data deluge scenario wherein more data is generated than can be
successfully/efficiently managed). Downsides of data deluge include
overlooking events due to alert fatigue (due to overwhelming data), missing
vital events, or insufficient time to process and leverage the data generated
by security tools.
Figure 4.3: Data deluge illustration
Let us examine SIEM solutions. SIEM is a powerful tool used for detection
purposes on a 24x7 basis thereby producing huge amounts of event data.
Unless there is an infrastructure that can support real-time integration (and
aggregation) of data, all this data may not be able to provide meaningful
inferences.
In a way, these tools create a lot of noise, which increases over time (e.g.,
from new endpoints and edge devices, cloud computing resources, etc.) as
there is no option for enterprises (businesses) to scale down data collection.
It is not easy for the security teams to be effective in these circumstances.

Challenge 2: Data analysis


In the above issue, we realize that sifting through the sheer volumes of data
created by cybersecurity tools is an issue. This data is also required to be
analyzed so that meaningful inferences can be drawn (refer to Figure 4.4
depicting the process of inspecting, cleansing, transforming, and modeling
data to discover useful information). To make this happen, data has to be
integrated from disparate and multiple sources. This is not an easy task by
any means.
Figure 4.4: Data analysis illustration
Let us take an example. We often see that network and application logs are
tracked separately. Anomalous data may come to the attention of a security
analyst in network logs, and he/she may not know its consequences. If
unusual activities are going on at the application level, they will not show
up at the network level. To decipher what transpires, the security analyst
must analyze the application logs from another cybersecurity tool.
This activity of correlation between network and application logs can be
difficult, rather slow, and prone to errors. The analysis is even more difficult
due to the volumes of data to be processed without appropriate tools.
Moreover, analysis needs to consider aggregating data across various
dimensions such as log severity level, time, resource type, and many more.
Query tools that work on integrated data sets are required to address this
challenge.

Challenge 3: Real-time security monitoring


In Chapter 1, Cybersecurity: A Dynamic Changing Paradigm, we
mentioned two aspects viz., Mean Time To Detect (MTTD) and Mean
Time To Remediate (MTTR), which are the ways to assess the overall
cybersecurity posture (performance) of an enterprise (business). It is
paramount for cybersecurity teams to find the source of an attack and
thwart the intrusion attempt before it can cause serious damage. For this to
happen, detection has to be swift, in real-time, and most importantly
cybersecurity teams must have access to the tools that can assist them in
taking action promptly and efficiently.
Refer to Figure 4.5, showcasing real-time monitoring which is the delivery
of continuously updated data about systems, processes or events.

Figure 4.5: Real-time security monitoring illustration


Attackers work at the speed of a machine. If they breach the defenses of an
enterprise (business) and inject malicious code into a system, they can
inflict huge damage. Human teams can ill-afford to respond to these
situations manually to block such attacks. The first preference is always to
block these kinds of attacks completely, and the second best is to detect the
events proactively and provide quick incident response.
Incident response depends on the incident type. An IP address can be
blocked if malicious code is detected in the traffic from that IP address. A
process can be terminated if it runs on a database server and data is being
exfiltrated in large amounts.
The actual response is not this simple. The process requires collecting data
regarding the attack and may involve monitoring the attack process to
understand its method. Overall, it is important to note that the incident
response has to be real-time (as much as possible) regardless of the action
taken (block or allow it to continue and monitor).

Challenge 4: Attackers think holistically


Malicious actors are motivated by the lure of lucrative potential revenues;
hence, they resort to acts like Ransomware and other cybersecurity attacks.
Cybersecurity attacks, especially Ransomware, lead to interruptions in the
physical world when they target critical infrastructures. Refer to Chapter 1,
Cybersecurity: A Dynamic Changing Paradigm, for details on various
Ransomware attacks. Figure 4.6 shows an attacker behind a camouflage,
ready to perpetrate an unauthorized action against computer infrastructure
to compromise it:

Figure 4.6: Attacker illustration


Countering such attacks requires a cybersecurity posture that can eliminate
inefficiencies and silos not only from an organizational standpoint but also
from within technology. This is precisely because hackers do not think in
silos. The limiting factor is that enterprises (businesses) often work in silos,
and so do various cybersecurity tools. The tools operate within their
specific views without or with bare minimum interoperability with the other
tools.
We have heard of lateral movement. In this, hackers use a weakness in a
system to exploit an adjacent area and can, move laterally within the
environment. This requires creating connections between security controls
through security analytics and security intelligence using domain-specific
information.

Challenge 5: Disjointed perimeter


Modern-day enterprises (businesses) do not have data and many
applications in their data centers (on promise). They are increasingly using
cloud-based applications that users access from anywhere. Figure 4.7
depicts a broken perimeter:
Figure 4.7: Disjointed perimeter illustration
In traditional circumstances, there would have been an on-premise data
center that would define the network perimeter security to control access.
The need of the hour is for context and identity to become the control
surface in a distributed environment where assets connect from anywhere.

Challenge 6: Adoption of multi-cloud strategy


According to various studies, enterprises (businesses) are moving towards a
multi-cloud setup since they take services from more than one Cloud
Service Provider (CSP). Since every CSP has a different set of policies,
creating a consistent security posture across multiple CSPs is not easy.
What adds to the challenge are the various services that are domiciled in the
on-premise environment.
Figure 4.8 illustrates a multi-cloud environment wherein an enterprise
(business) uses services from more than one public cloud provider at the
same time:
Figure 4.8: Multi-cloud strategy illustration
Having understood the various challenges being faced by today’s
enterprises (businesses), let us examine the benefits that CSMA presents to
address these challenges.

Benefits of CSMA
It is not adequate to meet today’s demands that are rapidly changing with
the existing approaches related to identity and cybersecurity architectures.
Existing approaches are quite disjointed. Therefore, the growing digital
landscape which is complex requires a new approach to cybersecurity
architecture. This is where the CSMA kicks in.
The basic objective of CSMA is to provide a common ground and a united
security framework to secure assets (on-premises and /or in the cloud).
CSMA enables point (stand-alone) cybersecurity solutions to work in an
integrated manner (also standardized) and bring about an improvement in
the overall security posture of the enterprise (business).
CSMA's further objective is to reduce the time required to collect and
organize security data, which can lend the enterprise (business) more time
to plan and respond to threats. Enterprises (businesses) need a platform that
performs this, which can be advantageous to IT/security operations. That is
just a baseline of what CSMA can assist with. Cybersecurity meshes can go
the extra mile by increasing visibility in the cybersecurity ecosystem,
reducing complexity in operations, and enhancing the sharing of security
intelligence across the enterprise (business).
Figure 4.9 illustrates a list of the various benefits of adopting CSMA:

Figure 4.9: Benefits of adopting CSMA

Enhancement of visibility
As discussed in Chapter 2, Cybersecurity: Understanding Today’s Security
Challenges, visibility is construed as being able to see and understand
various components of a cybersecurity infrastructure in a distributed
ecosystem. Also, in line with what is mentioned in the preceding paragraph,
CSMA follows an approach that is centered around the integration of
systems, thereby leading to increased visibility predominantly through
aggregation of security data and centralization of detection information.
The seamless aggregation permits the cybersecurity teams to distinguish
patterns easily in their data, which is the key.
Figure 4.10 portrays an eye to showcase how clearly various components of
a cybersecurity ecosystem can be seen:
Figure 4.10: Visibility illustration

CSMA gathers the output from various systems (e.g., by way of integrating
existing collected data and those from threat detection tools such as
Firewalls, access management/ identity, EDR, etc.) and further carries out
processing/analysis. This eventually leads to the end goal of CSMA, which
is to assist cybersecurity teams in gaining meaningful threat intelligence
(inferences).
CSMA also helps in the identification of patterns in security data that are
aggregated and can determine some of the attacks that may be well-hidden.
CSMA works to minimize the limitations of a distributed system where
security data is generated across various devices/tools, making it difficult to
correlate events. Let us learn this through an example. A cybersecurity
analyst observes an increase in usage of a computing resource on a server in
the cloud and would seem normal contextually. When seen in combination
with an IAM incident and event on the network, there could be a real cloud
security issue/event.
Let us reiterate what we have learned. Gaining visibility of data from
separate cybersecurity solutions is not adequate rather a cybersecurity
infrastructure (CSMA) is needed which enables cybersecurity teams to
better understand the connections between various components in the
ecosystem.

Reduction in complexity
In the case of cybersecurity, increased data aggregation usually goes hand-
in-hand with a reduction in overall complexity. Siloed security resources are
difficult to manage, as each section of an enterprise’ (business’) security
ecosystem will have different operational upkeep needs, and this kind of
structure can separate experts who should be communicating with one
another.
Figure 4.11 provides an expression of the state of being intricate or
complicated (aka maze):

Figure 4.11: Complexity illustration


CSMA reduces complexity by integrating siloed security systems into a
centrally observable and controllable operations dashboard. This
operational change lets SecOps staff respond to security incidents more
quickly and efficiently.
When suspicious activity is detected and communicated to staff through the
mesh architecture in an aggregated and integrated way, they are then able to
prioritize events more accurately. More information, and more centralized
controls, enable staff to exercise their security expertise by streamlining
security data collection and analysis.

Enhanced sharing of intelligence


Another important characteristic of CSMA is that it enables threat
intelligence and the sharing of intelligence across the cybersecurity
ecosystem. We have discussed previously that cybersecurity
tools/platforms/devices in siloed forms create a problem and lead to wasting
time and energy of the cybersecurity teams.
Figure 4.12 illustrates intelligence sharing which is the ability to exchange
intelligence, information, data, or knowledge:
Figure 4.12: Sharing of intelligence illustration
If implemented well, CSMA can enable an integrated view of the
cybersecurity ecosystem and permits seamless communication amongst
various resources across the enterprise (business) covering cybersecurity
experts by eliminating silos. This further leads to an increase in the
productivity of cybersecurity teams.
CSMA enables the sharing of intelligence amongst cybersecurity tools (on-
premises/cloud). CSMA rationalizes the collection of security data feeding
this aggregated data for enriching threat intelligence. Not only this, a
comprehensive CSMA will cover data (aggregated) analytics, the
application of machine learning, scoring of risk, and advanced analysis
tools. Overall, CSMA will add intelligence for policy enhancements and aid
in visualizations, orchestration, etc.

Enhanced automation
Attackers are known to be getting advanced with time and they are well-
synchronized. They always attempt to work in a stealthy manner such that
their attack goes unnoticed (rather than make it difficult to detect). We are
increasingly observing the use of attacks using AI. This makes it difficult to
respond to such attacks in time despite cybersecurity professionals working
overtime (shall we say, against time).
Figure 4.13 illustrates automation, the method of making a device, a
process, or a system operate by itself:
Figure 4.13: Automation illustration

The need of the hour for these professionals is AI-based automation which
can work across the cybersecurity ecosystem to adequately address the risks
posed by weaponization of AI.

Consolidation of security products


Of the many key priorities of cybersecurity, leaders are to consolidate
spending on security products/solutions and reduce complications in their
usage and management. According to a global report, enterprises
(businesses) on average use 45+ (refer to section Is the current situation
sustainable? of this chapter) cybersecurity products/solutions (Refer to and
technologies, which further means multiple vendors.
Figure 4.14 depicts consolidation, which is the action of combining several
things into a single more effective or coherent whole:
Figure 4.14: Consolidation illustration

In these circumstances, enterprises (businesses) need to have integrated


dashboards and management of interfaces centrally. In reality, no single
vendor offers all best-in-class cybersecurity controls (there is always one
better than the other). Therefore, often that enterprises (businesses) often
have to resort to using products/solutions from many vendors (also open-
source solutions at times). Consolidation into a smaller set of cybersecurity
vendors is a difficult task to achieve.
Supporting layers of CSMA are built through integrations. These
integrations can be brought about through multiple means viz., a mix of
proprietary APIs, open standards, and ad-hoc integrations between vendors’
tools. CSMA thereby offers an integrated view across the cybersecurity
ecosystem which assists the cybersecurity teams to respond swiftly and
effectively to security events.

Security analytics
We have observed that enterprises (businesses) use multiple cybersecurity
analytics tools together with other analytics tools. There are two problems
with this setup. Firstly, most of the analytics tools are domain-specific,
secondly, these tools do not work in a unified manner.
Figure 4.15 depicts a cybersecurity approach that uses data collection, data
aggregation and analysis tools for threat detection and security monitoring
(referred to as security analytics):
Figure 4.15: Security analytics illustration

As discussed earlier, CSMA provides a united security framework and


baseline support layers to secure enterprise (business) assets. Essentially it
enables otherwise disparate cybersecurity services to work in cohesion to
create a progressive and comprehensive cybersecurity posture. Due to this
unification of cybersecurity tools/solutions, CSMA enables the usage of
multiple cybersecurity analytics across cybersecurity domains.
This phenomenon has led to a trend/concept that we refer to as Extended
Detection and Response (XDR), which is much advanced in nature. Usage
of these lends enterprises (businesses) very mature cybersecurity
operations.
Using cross-domain cybersecurity data analytics and machine learning,
CSMA enables correlation at scale. This allows enterprises (businesses) to
detect patterns in data across cybersecurity sources, thereby identifying
more harmful threats. In essence, CSMA transforms observations into
intelligence that cybersecurity teams can act upon before events become
critical issues/disruptions.

Conclusion
We have observed that historically enterprises (businesses) have taken the
path of implementing best-fit solutions (that were appropriate for their
need). Examples of such solutions are mentioned in the section, The need
for CSMA. The limiting factor was non-interoperability among
cybersecurity tools which impeded the swift decision-making needed for
effective incident management. Inadequacy interoperability and
cohesiveness of cybersecurity tools/platforms results in increased overhead
in operations,
increased risk due to incomplete mapping of attack surface, incomplete
contextual information leading to ineffective decision-making and increased
incident’ MTTR and incident response.
As discussed in this chapter, CSMA leads to enhancement of visibility,
reduction in complexity, sharing of intelligence, enhanced automation,
consolidation of security products, and usage of security analytics across
domains in a complex and distributed cybersecurity ecosystem. This results
in driving a larger Return on Investment from the existing investment
(which is the most difficult variable to draw out in cybersecurity),
prioritizes risk based on business impact, and most importantly enables
effective incident management and response through intelligence sharing
(across domains).
As a concluding remark, we should say: There is all the benefit in
cybersecurity meshing.
In the subsequent chapter, we shall further examine the key components of
CSMA, discuss the outcome of the adoption of CSMA, and conduct a sneak
preview of CSMA products/solutions.

Points to remember

With the adoption and growth of Cloud and hybrid (decentralized)


working, cyber ecosystems have become fragmented
Securing yesterday’s distributed security architectures is a difficult task
This is where cybersecurity mesh comes into play. It is a new
architectural approach that reduces the need to have one specific
computing environment
CSMA is an approach that allows the integration of security tools into
a scalable and collaborative ecosystem. It aims to create a much more
stable and reliable security as compared to traditional distributed
security architectures
CSMA is a concept than a concrete end-to-end solution, however is a
viable framework that holds analytics, controls, and threat hunting to
its core
CSMA works towards creating a decentralized cybersecurity
ecosystem, where security tools are better aligned to complement each
other and integrate across all network components

Key terms

CSMA: A contemporary scalable approach to cybersecurity


architecture that allows the distributed enterprise (business) to deploy
and extend security controls where it is needed the most.
NAC: The process of restricting unauthorized users and devices from
gaining access to a corporate/private network.
IPS/IDS: Systems that monitor the network and identify potential
incidents (also log and report security incidents). IDS systems do the
detection, while IPS tools work toward prevention.
EDR: A set of tools used to detect and investigate threats on
endpoints.
API: A software in-between two applications that permits the two to
talk to each other.

Join our book’s Discord space


Join the book's Discord Workspace for Latest updates, Offers, Tech
happenings around the world, New Release and Sessions with the Authors:
https://discord.bpbonline.com
CHAPTER 5
Fundamental Components of
Cybersecurity Mesh Architecture

Introduction
In the previous chapter, we discussed the concept of CSMA, its need for
enterprises (businesses), and its benefits in detail. A security mesh
architecture is a philosophy and not necessarily a single end-to-end security
solution. Some fundamental aspects of CSMA, such as its key components
and the merits of its adoptions by enterprises (businesses) will be discussed
in this chapter.

Structure
In this chapter, we will cover the following topics:

A re-look at CSMA
Key components of CSMA
The unified architecture of CSMA
Major products/solutions of CSMA

Objectives
This chapter will discuss various layers (referred to as key components)
that make up a CSMA, the interaction amongst these layers, and how this
architecture contributes to a progressive enhancement of the Cybersecurity
maturity posture of an enterprise (business). We shall further discuss the
outcome of the adoption of the unified architecture as propagated by CSMA
and have a look at CSMA products/solutions across various categories.
A re-look at CSMA
Layers of vulnerabilities and gaps due to the absence of interoperability of
security tools/solutions are forcing security models to be redefined. Added
to this equation is the ever-growing complexity of cloud (multi and hybrid)
environments that is changing the threat landscape. Refer to section, The
current situation of the cybersecurity ecosystem of Chapter 4, The
Need for Cybersecurity Mesh Architecture, wherein most enterprises
(businesses) use, on average, 45+ cybersecurity tools/solutions, which leads
to unmanageable complications and maintenance overheads to manage each
tool/solution.
CSMA addresses these issues by using modular technologies (also scalable
and flexible) along with strong standards and policies. This creates an
ecosystem of security tools/solutions that integrate across all its components
and provide maximum security.
Another way of looking at this is the fact that each computing environment,
such as Infrastructure as a Service (IaaS), Software as a Service (SaaS),
virtual machines, endpoints, etc., is generally siloed (which means security
controls are specific to the environment). CSMA enables a security posture
through the integration of tools/solutions.
In effect, CSMA aids enterprises (businesses) to move towards an
integrated ecosystem from a siloed/standalone philosophy. It provides a
framework that revolves around analytics, controls, and threat hunting and
follows Zero Trust objectives.
The benefits of implementing CSMA are depicted in the following
illustration:
Figure 5.1: Key benefits of CSMA

CSMA working relationship with zero trust


We have heard of zero trust many times over. Both CSMA and zero trust
aim to improve the cybersecurity posture of enterprises (businesses) but
have different areas of focus. Therefore, it is safe to state that CSMA does
not replace zero trust or vice versa.
CSMA aids in the improvement of an enterprise’ (business’s) cybersecurity
at an architectural level by removing silos and enabling a granular level of
security. It improves the efficiency of security threat detection and incident
response capabilities.
Whereas, zero trust is a philosophy of cybersecurity that needs to be
implemented within a CSMA. There are some critical requirements of zero
trust, such as principles of identity, authentication, access control, micro
segmentation, consistent security enforcement, etc. which make up for an
efficient building of CSMA.
Thus, one of the critical foundations of CSMA is zero trust, and CSMA is a
wrapper for enterprises (businesses) to embrace zero trust through flexible
and scalable services/tools/processes.

CSMA’s scalability relative to other architectures


Today's security architectures comprise tools/solutions that seamlessly do
not work together, resulting in security gaps. Consider that you are putting
together a building using interconnecting blocks. You can add pieces
anywhere as part of the design, or you can take out some from a specific
spot if it is not required there. You have the liberty to add or remove pieces
in certain circumstances. Moreover, all the pieces work together under all
circumstances.
This is known as modular (composable) technology and is the foundation
of CSMA. This makes CSMA scalable and flexible. Such an architecture
permits resources (real/virtual) to meet computing needs as they arise.

Working of CSMA
A vital principle of CSMA is to combine a set of computing tools/resources
using a unified application programming interface (API). This allows the
pool of available resources to be requested automatically as required for
particular applications/workloads.
In CSMA, processes take an API-first approach, wherein programmers
write software code after conceptualizing how an API will behave and
address specific business requirements. CSMA incorporates regimented
standards and security processes in the initial connections not only among
applications but also between hardware and software.
When this happens, hardware becomes a secondary issue, and the
connections as the process drivers become more secure and take
precedence. Also, due to the usage of unified API, the new connections
combine AI, M, and automation to make the enterprise (business) security
scalable and less complex.

Key components of CSMA


CSMA has the following key components:

Integrated data
Distributed identity fabric
Comprehensive security intelligence
Automated operations
Centralized operations
Consolidated dashboards

The central themes of CSMA are modularity, scalability, and collaboration.


These enable disparate security controls to work collaboratively and ease
their configuration and management.
Figure 5.2 depicts the key components of CSMA:

Figure 5.2: Key components of CSMA

Integrated data
This is the foundational component of CSMA. Integrating data from
disparate sources across security systems is vital. There are distinct
advantages to doing so. First, this enables security solutions/tools/products
to talk to each other, which enhances each component’s effectiveness and
flexibility. Second, data analysis and data mining (which enhances
intelligence) also become effective and meaningful.
For example, a security alert may be received from an IAM tool indicating
an inappropriate access attempt or an account being created incorrectly.
This could seem like an important notification, but there could be a
weakness in the IAM tool, which can only be known with the assistance of
other security tools/systems.
In the case of CSMA, a cloud monitoring tool may be able to correlate and
associate unusual access to a cloud computing resource with a privileged
account. This may lead to questioning the IAM tool for more information.
Therefore, the data integration component in CSMA is very powerful. It is
not just data aggregation; it enables siloed security systems to work much
more effectively through multi-event complex correlations.

Distributed identity fabric


This is a critical component that deals with IAM functions, such as
directory services, management of identities that are distributed, adaptive
access, authentication, authorization, and entitlement management
capabilities suitable for today’s distributed environments.
The need of the hour is for each verified identity to access the resources it
needs from devices that are allowed and locations that are approved.
Traditionally, in standalone architecture setups, IAM functions are generally
managed by different businesses, and this has the potential to leave security
gaps. In CSMA, policy regarding identities is centrally managed and,
therefore, applied consistently. This leads to better management of risk,
lends more trust, more security, and enhanced user experience.
As an example, a Cloud Access Security Broker (CASB) continuously
monitors what users are performing and will use the identity fabric to
prompt the access control tool for MFA when appropriate. CSMA makes
IAM tools more composable to manage the new IAM use cases.

Comprehensive security intelligence


The next component of CSMA is comprehensive security threat intelligence
(and data analysis) on a real-time basis. This aspect deals with the
processing of aggregated security data from disparate sources that enable
cybersecurity teams to make meaningful and informed decisions. This
further provides impetus to conducting data analytics to improve the overall
threat intelligence for the enterprise (business). Threat intelligence is
generally driven by technologies like machine learning (ML) and
artificial intelligence (AI).
Many solutions operate at this level, such as:

Security Information and Event Management (SIEM)


Security Orchestration Automation and Response (SOAR)
Extended Detection and Response (XDR) and
User and Entity Behaviour Analytics (UEBA).

CSMA involves integration between solutions (such as above), enabling


access to analytic tools in a central place for enterprises (businesses). This
provides simplicity to both management and operational-level activities.

Automated operations
The previous components focused on integration and analysis. This one
works towards streamlining management of 2Ps viz, Policy and Playbook.
This is, therefore, the management component of CSMA and enables the
setting up of policies. These policies govern how and when alerts are
triggered and how data is processed for analytics. CSMA deals with
centralization and integration and enables the setting up of policies for any
component connected to the security mesh.
Whether it is SIEM logs, zero trust policies, or IAM controls, all can be
configured from a central place. Similarly, event playbooks can also be
integrated into security systems, lending a comprehensive security incident
response.

Centralized operations
This component of CSMA covers dashboards and operational controls
under one roof. A CSMA dashboard takes time to build and includes tools
for investigating and reporting events and visualization tools.
Visualization tools can especially be utilized to generate a comprehensive
risk score of the complete security stack in a mesh (this is only possible
because of holistic security data aggregation).
Consolidated dashboards
The last component of CSMA deals with the overall security posture of an
enterprise (business). Traditionally, in a standalone/distributed/siloed
environment, security posture would be examined through specific
dashboards, for example, an EDR dashboard for endpoints or a Cloud
Security Posture Management (CSPM) dashboard for Cloud.
With CSMA, teams can have a unified (and consolidated) view of the
security posture of all assets (on-premise/cloud) centered around identities
in the form of dashboards. This enables swift and agile security incident
response.
In summary, in CSMA, standalone/distributed/siloed solutions work
together in a complementary (interconnected) manner, improving the
overall security posture of the enterprise (business).

CSMA’s most important component


Of these key components, we should mention the identity fabric is special
because it enables security, which is driven by and based on identity.
Identity is the most vital element in the security stack, as it is known to be
the new edge (network), particularly in the cloud (where controls on
identity are used to protect sensitive data).
In CSMA, let us remember that IAM systems provide context about events,
which proves that identity is core to security.

The unified architecture of CSMA


CSMA covers a wide spectrum of integrated security capabilities across the
cybersecurity ecosystem. It uses AI to power automation in order to
improve an enterprise’ (business’) security posture in the wake of complex
modern cyber threats.
Figure 5.3 depicts the core philosophies of CSMA:
Figure 5.3: Core philosophies of CSMA
Let us now examine the core philosophies of CSMA, which progressively
aid enterprises (businesses) in maturing their cybersecurity posture.

Extensive
A regimented CSMA needs a spate of integrated analysis tools to aggregate
security data across the cybersecurity ecosystem. Please note that CSMA
needs sufficient data to work and abstract the operations of the network,
endpoints, and cloud security solution assuredly. enterprises (businesses)
need a wide range of superior-quality security systems to support the
implementation of CSMA.
For example, solutions such as zero trust access, network security, cloud
security, etc., should have the required communication capabilities that
would enable them to integrate into the larger cybersecurity ecosystem.

Integrated
CSMA includes an array of cohesive analysis and response mechanisms
that help connect various components of a distributed IT/cybersecurity
ecosystem. As mentioned earlier in this chapter (Refer to the section
Comprehensive security intelligence), solutions such as SIEM, SOAR,
EDR, etc., come under this roof. Moreover, each solution integrates with
the cybersecurity ecosystem and provides the much-required threat
intelligence for swift and efficient incidents.
CSMA takes time to build and progressively mature over time. It is
intricate, which is evident from the fact that various features and services go
into it, and is very flexible (scalable). Each part of the cybersecurity
ecosystem can be tweaked over time to meet the needs of the enterprise
(business).
The key to managing this goal is the integration of cybersecurity solutions.
The underlying unified security architecture is thus able to reduce the
complexity in the management of cybersecurity solutions/tools while
enhancing threat intelligence.

Accelerated through automation


The best way to lessen operational overheads and increase the effectiveness
of cybersecurity teams is through automation. We have discussed that
automation is one of the key components of CSMA.
Automation (end-to-end) in detection, prevention, and threat hunting is vital
to defend against a real-world cyberattack (coordinated). This enables
enterprises (businesses) to proactively monitor, identify, and thwart cyber
threats such as ransomware, phishing, other advanced attacks, etc.
It is equally important that policies for cybersecurity services in CSMA are
managed centrally such that they can be executed swiftly and consistently
when an attack occurs.
One word of caution here. There are claims that vendor products include AI
and ML capabilities to enhance automation, but remember, AI and ML are
as good as the data they are trained on and the people who train them.
Enterprises (businesses), therefore, need to carry out their due diligence
before picking up new technologies. CSMA aims to ensure that automation
is trustworthy.

Major products/solutions of CSMA


Various products/solutions are required to satisfy the requirements of
CSMA. Their products/solutions aid in closing the gaps and mitigation of
risks.
Here is a list of major ones under categories:
Figure 5.4 lists the CSMA solutions that belong to the Network category:
Figure 5.4: CSMA solutions belonging to Network category

Zero Trust Network Access (ZTNA)


Intrusion Detection and Prevention System (IDPS)
Secure Web Gateway (SWG)
Enterprise Firewalls (EFW)

Figure 5.5 lists the CSMA solutions that belong to the Email category:

Figure 5.5: CSMA solutions belonging to Email category

Secure Email Gateway (SEG)

Figure 5.6 lists the CSMA solutions that belong to the Identity category:
Figure 5.6: CSMA solutions belonging to Identity category

Security Information and Event Management


Identity Governance and Administration (IGA)
Access Management (AM)
Privileged Access Management (PAM)
Customer Identity and Access Management (CIAM)
Multi-Factor Authentication (MFA)

Figure 5.7 lists the CSMA solutions that belong to the Data category:
Figure 5.7: CSMA solutions belonging to Data category

Data Classification
Enterprise Digital Rights Management (EDRM)
Data Loss Protection (DLP)

Figure 5.8 lists the CSMA solutions that belong to the Applications
category:
Figure 5.8: CSMA solutions belonging to Applications category

Web Application and API Protection (WAAP) / Web Application


Firewall (WAF)
Dynamic Application Security Testing (DAST)
Static Application Security Testing (SAST)

Figure 5.9 lists the CSMA solutions that belong to the Devices/Servers
category:

Figure 5.9: CSMA solutions belonging to Devices/Servers category

Endpoint Detection and Response (EDR)


Extended Detection and Response (XDR)
Endpoint Protection Platform (EPP)
Unified Endpoint Management (UEM)/ Unified Endpoint Security
(UES)

Figure 5.10 lists the CSMA solutions that belong to the Cloud category:
Figure 5.10: CSMA solutions belonging to Cloud category

Cloud Workload Protection Platform (CWPP)


Cloud Access Security Brokers
Cloud Security Posture Management (CSPM)

Conclusion
Through this chapter, we understood that implementation of CSMA results
in enhanced agility, flexibility, scalability, and cybersecurity posture.
CSMA takes an enterprise’ (business’) cybersecurity infrastructure to an
elevated level owing to the integration of various cybersecurity services. All
communications coming in and going out of the network are protected,
making it safer. A message to all enterprises (businesses) is to be future-
ready. Choose cybersecurity solutions that facilitate integration. Use plug-in
APIs to aid customization and interoperability (further aiding data analytics
and threat intelligence).
Bridge security gaps due to weaknesses/vulnerabilities, if any, in
cybersecurity solutions using current and enhanced CSMA security
standards.
Implement all key components of CSMA by laying emphasis on all key
components, such as data integration, identity fabric, policy management,
security analytics, threat intelligence, and integrated dashboards. A
combination of all components (aka overarching solution), provides an end-
to-end and a holistic cybersecurity ecosystem. In the next chapter, we shall
further into the unified architecture of CSMA, discuss the importance of
integration, comprehensiveness, and automation in CSMA, and examine the
key factors to consider while adopting CSMA.

Points to remember

CSMA creates a way for distinct/standalone cybersecurity services to


communicate and integrate.
CSMA enhanced the cybersecurity posture of an enterprise (business)
by making it more agile, scalable, and flexible.
CSMA facilitates enhanced collaboration, analytics, and threat
intelligence thereby greatly enhancing incident response.
The key foundational layers of CSMA, viz., Integrated data,
distributed identity fabric, comprehensive security intelligence,
automated operations, centralized operations, and consolidated
dashboards, provide a roadmap for implementation.
CSMA and zero trust should not be used interchangeably.
While CSMA focuses on granular and integrated security, the zero-
trust principle removes implicit trust and provides visibility and
control over each access request performed within its ecosystem.
Zero trust is one of the critical foundations of CSMA.

Key terms

Zero Trust Network Access: Technology and functionality set that


enables secure access for remote users to internal applications.
Secure Web Gateway: Protects enterprises (businesses) from online
security threats by enforcing policies and filtering Internet traffic. It is
available on-premise or via cloud
Enterprise Firewalls: Controls network traffic between untrusted
sources (e.g., public Internet) and trusted sources (e.g., the private
enterprise network).
Secure Email Gateway: Security solution used for emails that filter
emails for suspicious and potentially malicious content.
Identity Governance and Administration: Facilitates security
administrators to manage user identities and access across the
enterprise (business) efficiently.
Access Management: Practices that enable an entity to act on a
particular resource when explicitly permitted.
Privileged Access Management: Security solution that protects an
enterprise (business) against cyber threats by monitoring, detecting,
and preventing unauthorized privileged access to critical resources.
Customer Identity and Access Management: Enables enterprises
(businesses) to capture (securely), manage customer identity/profile
data, and control access to applications/services.
Enterprise Digital Rights Management: Controls and manages
access to enterprises’ (businesses’) copyrighted material.
Data Loss Protection: Cybersecurity solution to detect and prevent
data breaches (through leakages).
Web Application and API Protection: Cybersecurity technologies
aimed at protecting web applications and APIs from sophisticated
cyberattacks.
Web Application Firewall: Protects web applications through
monitoring and filtering HTTP traffic between the web application and
the Internet.
Dynamic Application Security Testing: A black-box testing to find
out vulnerabilities in web applications from outside of the enterprise
(business).
Static Application Security Testing: A white-box testing that finds
out vulnerabilities inside the application and code.
Extended Detection and Response: XDR combines data from
multiple security products, including EDR, network, cloud, and email
security, to provide a holistic view of security threats across the
enterprise (business).
Endpoint Protection Platform: Cybersecurity solution built to detect
and block threats at the device level.
Unified Endpoint Management/Unified Endpoint Security: UEM
solutions carry out general management of endpoints at scale. UES
delivers some management capabilities however, it focuses on security.
Cloud Workload Protection Platform: Security solution or cloud that
protects cloud workloads in a multi-cloud/hybrid environment.
Cloud Access Security Brokers: Provides security policy
enforcement between CSP and consumers (can be on-premises/cloud-
based).
Cloud Security Posture Management: Security tools built to find out
misconfiguration issues and compliance risks in the cloud.

Join our book’s Discord space


Join the book's Discord Workspace for Latest updates, Offers, Tech
happenings around the world, New Release and Sessions with the Authors:
https://discord.bpbonline.com
CHAPTER 6
How to Effectively Adopt
Cybersecurity Mesh Architecture

Introduction
In the previous chapter, we discussed the fundamental components of
CSMA, which enable it to deliver a strong framework for security
infrastructure. While the modern digital world witnesses advanced threats
that are constantly evolving, various state-of-the-art edge security tools rise
to the occasion to address this challenge (but in a highly distributed
environment). CSMA takes this response to the next level by integrating
these resources.
In this chapter, we will discuss what to expect when adopting a CSMA and
how enterprises (businesses) can effectively prepare for the future.

Structure
In this chapter, we will cover the following topics:

Performance
Cybersecurity landscape of today
Key aspects for CSMA adoption
Getting started with CSMA
Key factors to consider while adopting CSMA

Objectives
We discussed the key foundations of CSMA in Chapter 4, The Need for
Cybersecurity Mesh Architecture. For new enterprises (businesses), these
key foundations present a blueprint for designing and building the
cybersecurity architecture. However, this is challenging for many
enterprises (businesses) due to the existing legacy of the IT ecosystem
comprising various security solutions.
No one vendor can provide all the required building blocks, comprehensive
standards are in the works, and cybersecurity products/services do not
generally interoperate. The objective of this chapter is to walk through the
steps enterprises (businesses) can undertake to adopt a comprehensive
CSMA.

Performance
Let us examine the treatment of cybersecurity challenges thus far.
Standalone/siloed cybersecurity solutions have a tradition in dealing with
these challenges. Enterprises (businesses) have thrown standalone solutions
for endpoint security, cloud security, network security, email security, etc.,
in the fold and try to deal with the challenges.
This is unfortunately not workable. Modern-day IT infrastructure
operational needs are complex. For example, IoT, hybrid clouds, an
employee using more than one device, and offices that are geographically
spread and separated are increasingly complex (resources are widely
distributed).
There is no way this ecosystem will change because this is lending
effectiveness and competitiveness to enterprises (businesses).

Problem with past solutions


While it may feel satisfying (the most suited course of action) to have a
best-fit security tool for an individual component of the IT ecosystem, this
treatment of securing the IT ecosystem misses something critical.
Large and mid-sized enterprises (businesses) generally have resources
spread across multiple geographies and technologies, which means it is not
a matter of securing each component of the ecosystem infrastructure, but
what matters the most is how to secure all of the components
simultaneously and effectively.
Best-fit security solutions/tools may be reasonable answers to specific
security challenges, but that is now how the modern IT ecosystems are put
together.
Modern-day IT ecosystem is about the co-existence of networks, end-
points, cloud resources, and edge devices. They do not exist or work in
isolation. These components are dependent on each other to function
properly; therefore, they need to be secured together.
It is important to look at these components in an integrated manner. For
example, advanced firewalls may get breached and there has to be a way to
swiftly determine the other areas where an attack might have been
perpetrated.
This is a known constraint of standalone/siloed security solutions. Their
ability to communicate with each other is limited, which further hinders
their capability to detect. Also, the potential for responding to security
incidents and breaches gets impaired.
When IT ecosystems were simpler (with fewer IT components and more
centralized), standalone/siloed security tools/solutions worked well. Siloed
configurations are much less effective now, with complexities rising in the
IT ecosystem.
When security tools/solutions do not talk to each other, security teams have
less visibility of the attack surface. This has an impact on the teams as they
get separated from each other, thus not benefiting from interactions with
other experts.
Please remember that the security and operation teams in an enterprise
(business) that interact with each other work more effectively. Therefore, it
is imperative that security staff work on problems centrally, where holistic
and comprehensive security controls enable them to adjust to newer cyber
threats.

Answer lies in the unified architecture


In the past, we have seen the emergence of various security solutions, such
as SOAR, which work to unify the security infrastructure. Though they
have added value to many enterprises (businesses), they are missing the
more extensive foundation that endorses philosophies of integration and
aggregation of security data. It is vital for a successful ecosystem nowadays
to have integrated visibility.
Attackers have a reasonable understanding of the attack surface of an
enterprise (business) and, thus, know how to take advantage of weaknesses
due to this awareness. It is the element of security tools’/solution’s data
generation to be centrally visible that a CSMA brings to life.
We have also realized that automation has been a savior to enterprises
(businesses) and cyber attackers who utilize these tools to maneuver
through compromised systems swiftly.
Two aspects are best suited to counter these evolving threats: security
controls that are centralized and tools that are automated. Cybersecurity
experts should solve complex problems where possible and not waste their
precious time shutting down network ports to thwart a threat.
Another challenge discussed in the previous chapters is the lack of security
talent. While the cybersecurity experts already on board should be
leveraged to increase the security posture, CSMA makes automation of
security and centralized policy management possible.
With the visibility of all vital security components and being centrally
managed, security and operation teams can configure, detect, and reassess
swiftly.

Cybersecurity landscape of today


The cybersecurity landscape today is like a game of table tennis between
cybersecurity professionals and cyber attackers.
Exploring the game reveals that types of cyberattacks keep evolving (some
attacks keep falling out, and others emerge), and cybersecurity experts, in
response, keep innovating new defense techniques to counter this challenge.
Figure 6.1 depicts the top challenges for the adoption of CSMA:
Figure 6.1: Top challenges for CSMA adoption

Attack surface expansion


The post COVID era has witnessed a major shift from work-from-office to
hybrid working (owing to the push to work from home). Employees are
increasingly taking corporate devices to their homes, connecting them to
home networks, thereby becoming more vulnerable to attacks on the
endpoint device/phishing attacks as compared to a situation when they were
working from office premises.
The cybersecurity and operation teams need holistic visibility to observe
how and when attacks take place, especially threats where it is not easy to
identify sources. Let us consider a few examples in this context:

An attack emerges from an endpoint device. As the attack surface


expands, it becomes harder to determine the source of such an attack.
An attack is noticed by way of network detection. This necessarily
does not mean that the network is the source of the breach. A
distributed IT ecosystem makes it even more difficult to investigate
and analyze such an attack.
An attack dealing with the exfiltration of data from a database. This
could render cybersecurity teams clueless about its source. A siloed
security system would have difficulty finding the source of the
problem, not to mention other potential points of infiltration.
We know that analysis begins after a threat is detected and remediated.
Even though cybersecurity teams become knowledgeable regarding system
weaknesses through analysis, a lack of comprehensive visibility can impede
investigations.
Modern-day IT ecosystems need cybersecurity tools that can carry out
comprehensive analysis after an attack, and a CSMA enables this.

Adoption of cloud
Over the last few years, enterprises (businesses) have been working to
move their workloads to the cloud, predominantly to utilize the benefits of
cloud technologies. However, one common concern among everyone is the
security concerns surrounding cloud adoption. Two common concerns in
this light are inadequate visibility and the absence of comprehensive
security controls.
We have seen how complex an IT ecosystem is. The adoption of the cloud
adds a layer to this complexity. What increases the security risk are the
following aspects: new entry points for attackers, new tools, and
unfamiliarity of staff (administrative/engineering) with new technologies.
If the lack of visibility and absence of comprehensive controls is added to
this mix, the challenge is further compounded. If this is not enough, many
enterprises (businesses) work in a hybrid/multi-cloud mode where
technologies differ, further adding to the complexity.
CSMA is a method of addressing security challenges in the cloud by
aggregating security data from a wide variety of security resources and then
centralizing security controls to ensure ease of use.
Security tools need integration and aggregation in the future, and CSMA is
a flexible and scalable method to commence nurturing that kind of security
ecosystem.

Key aspects for CSMA adoption


Truly modern security systems are more than just tools to meet today’s
cybersecurity threats. They are collections of methods and solutions that
meet those threats while allowing for flexibility in the ever-changing
security landscape.
CSMA enables the creation and maintenance of dynamic security systems
while also adding powerful integration and aggregation capabilities that
promote a strong security posture.
In this section, let us review some of the key points of cybersecurity meshes
that make it a strong, modern option for hardening your organization’s
resources against both old and new attacks.
Figure 6.2 lists the key aspects for adoption of CSMA:

Figure 6.2: Key aspects for CSMA adoption

Integration
Amongst many vital aspects of a strong cybersecurity ecosystem are the
visibility of data and infrastructure. This helps combine security resources
centrally, thus aiding in preventing hackers from operating in a stealth mode
and exploiting a weak spot.
CSMA works to create connections among security tools/solutions that
were otherwise standalone/siloed and enables aggregation of security data
for easier analysis and observability.
A holistic data collection capability also allows the system to learn from
past cyberattacks and progressively improve the cybersecurity posture.
CSMA enables the integration of security data sources. This supports
security and operation teams in swiftly and efficiently collecting and
analyzing information after an incident.
Similarly, visibility across the cybersecurity ecosystem in an integrated
manner, with near- almost real-time updates, enhances the speed of
detection and response.
Therefore, enterprises (businesses) need to address the absence of
integrated visibility and lack of central control to maintain and secure a
multi-cloud environment.

Comprehensiveness
We just referred to visibility as one of the vital aspects of a strong
cybersecurity ecosystem. However, visibility does not have meaning if you
cannot do much with it. Equally vital is for the cybersecurity ecosystem to
be flexible (scalable) and efficient through strong control capabilities.
CSMA works towards centralizing security controls (as much as possible)
by aggregating data and creating comprehensive dashboards (central
consoles for efficient control). The benefits of central access are efficient
reporting, swift investigation, and real-time alerting.
CSMA does not only lead to the integration of security data but also lends a
centralization theme to control the IT ecosystem. Orchestration and policy
management are centralized in a CSMA.
Policies are not required to be set on a one-on-one basis for a wide variety
of security tools/solutions because, in a CSMA environment, tool/solution
functionality is integrated.

Automation
When cybersecurity tools/solutions are integrated into a control system
centrally, automation of security tasks becomes effective and powerful. It is
not only cyber attackers who use automation to swiftly move laterally
through systems to carry out complex attacks. Automation is equally useful
for defense tools in the present circumstances (and will be in the future as
well).
CSMA enables a variety of tasks to be automated with great flexibility.
Device configuration tasks can be set up centrally, so they can be
notifications/triggers freeing up the time of security and operation teams.
Overall, CSMA is built to weave automation of tasks through cybersecurity
tool/solution integration (by leveraging AI and ML) that enables dynamic
policies and configurations.
Getting started with CSMA
The key components of CSMA, as discussed in detail in Chapter 4, The
Need for Cybersecurity Mesh Architecture, present a blueprint roadmap for
designing, building, and implementing CSMA.
This is true for a greenfield setup, but for many enterprises (businesses), the
start will be from a legacy IT ecosystem—a complex ecosystem with many
cybersecurity tools/solutions already in place.
Many challenges need to be overcome for a successful adoption of CSMA.
For example:

A single vendor does not have all the building blocks required for
CSMA
Standards required for CSMA are still in the works
Tools/solutions do not yet provide the required interoperability

Despite this, the following steps shown in Figure 6.3 can help enterprises
(businesses) advance toward an effective implementation of CSMA
implementation:

Figure 6.3: Key steps to get started with CSMA


Assess current tools and identify gaps
As for any cybersecurity program, the first step is to review your existing
cybersecurity strategy, risk profile, tools/solutions, controls, and skillsets.
The need is to comprehensively map out the data flows between various
tools/solutions.
The vital next step is to assess the maturity of the current tools/solutions
based on their functionality and integration capability. Seek out the gaps
leading to white spaces (blind spots) in your infrastructure. For example,
are there missing risk signals about device status or an inability to track a
user’s activity (on-premises/on the cloud)?

Build the key components


Next, commence the development of key components (refer to section Key
components of CSMA in Chapter 4, The Need for Cybersecurity Mesh
Architecture) as follows:

Data integration:
This is a powerful component in CSMA.
Evaluate tools/solutions that support data aggregation enabling
effective working of siloed security systems with the ability to
perform multi-event correlations.
Distributed identity fabric:
This is the most critical component.
Seek tools/solutions that assist in the central administration of
identities and can strictly enforce access policies (sophisticated)
across the IT ecosystem.
Comprehensive threat intelligence (and analytics):
Tools/solutions such as SIEM, UEBA, SOAR, XDR, etc., that can
integrate well with the core cybersecurity and identity solutions.
Provide risk scoring that is dynamic and entity-based.
Automated and centralized operations:
Look for tools/solutions that can streamline the management of
2Ps, such as policy and playbook.
Have the capabilities aligned with NIST, CIS, and ISO (globally
recognized and established security standards).
Consolidated dashboards:
Works towards an ideal dashboard that will provide clear
visualizations of risk scoring and alerts on a real-time basis and
comprehensive insights.
Decide whether you are seeking multiple views (for different IT
security and stakeholder roles), as well as advanced features such
as customizable widgets and reports.

Evaluate compatible tools


Enterprises (businesses) need to choose security tools/solutions and services
capable of deep integration. As mentioned earlier, standards for CSMA are
still in the works, but some of the current examples are as follows:

OAuth 2.0
OpenID Connect
System for Cross-domain Identity Management (SCIM)
Open Policy Agent (OPA)

Equally important is the principle of open APIs, which enables


cybersecurity tools/solutions to share information seamlessly, including the
context of identity and risk intelligence.
For example, if we want the network firewall and email gateway to be able
to make decisions regarding authentication, the two entities should be able
to communicate with each other first and foremost.

Consider befitting vendor(s)


So far, we have emphasized the choice of cybersecurity tools/solutions.
Choosing a quality vendor(s) is as vital as the tool/solution. Suitable
vendors with extensive experience and a proven track record of being
receptive to changes in the cybersecurity landscape and adjustable in
assisting customers secure their ecosystems.

Seek consolidation
Post-evaluation of suitable vendor(s) and cybersecurity tools/solutions,
enterprises (businesses) should look to simplify their cybersecurity stack
and licensing regime. This can be achieved through vendor and
tools/solutions consolidation.
Many enterprises (businesses) in today’s world have already commenced
the journey of cybersecurity vendor consolidation.
This level of consolidation should be at an appropriate level. If an enterprise
(business) is seeking one identity provider to limit its identity fabric, it may
not be reasonable to effectively manage all the use cases.
There are various internal employee groups, external contractors,
customers, partners, etc. All these entities have specific identity use cases
that will be difficult to manage.
It is advisable to have a CSMA vision with an underlying cybersecurity
roadmap of tools/solutions/services. The objective should be to have an
integrated ecosystem of tools and standards that does not lead to elevating
your technical debt in the long run.

Review and revise


Designing and implementing CSMA is not a one-time activity; rather, it is
an ongoing project. It cannot be static since everything around the
enterprise (business) is continuously changing, viz., business requirements,
standards, technologies, available tools/solutions/services, threat landscape,
etc.
It is vital to evaluate the performance and effectiveness of your
cybersecurity posture in a regimented manner. This will enable you to make
adjustments as required.
Here, you should specify the metrics that are required to be tracked and
reviewed regularly and measure how a CSMA strategy is having an impact
on business outcomes.
The key is to monitor areas where weaknesses exist. These can include
tools/solutions/services that are deficient in functionality, do not integrate
well with other entities in the ecosystem, and are not adequately supported
by vendor(s).
Also, as a matter of agility (timely action), keep an eye on product
enhancements. Any increase in the maturity of a tool/solution can present
an opportunity for consolidation which can be a boon to simplifying the
technology stack.

Key factors to consider while adopting CSMA


We have seen many advantages associated with the adoption of CSMA, but
let us now explore reasons why enterprises (businesses) are considering
going this route.
Figure 6.4 lists the factors for considering CSMA:

Figure 6.4: Factors for considering CSMA


Let us look at the following factors:

Vulnerability:
IT ecosystems are vulnerable to phishing and ransomware attacks,
both of which have been on a steady rise year-on-year.
If this is not difficult enough, there exists the danger of a zero-day
attack (explained in the previous chapters, this is an attack by a
method that was unknown previously).
CSMA enables an enterprise (business) to reduce these
vulnerabilities significantly and prepares its cybersecurity
infrastructure for such attacks.
Cost:
Global ransomware damage costs will reach $20 billion by 2021
(57x more than it was in 2015), and the cost of cybercrime attacks
is on the rise at about 15% per year over the next five years1.
There is a huge cost of digital transformation as well for
enterprises (businesses) whose infrastructure/architecture has to
undergo redesign.
It is reasonable to state that the reduction in attacks (costly) and
downtime due to leveraging CSMA far outweighs any initial cost.
Migration:
Many clients/consumers have shifted to cloud computing owing
to their benefits, which has led to an increase in data breaches.
CSMA enables migration to cloud computing by providing
flexible and scalable protection.

The above-mentioned factors showcase the need for CSMA. Figure 6.5 lists
the add-on factors for considering CSMA:
Figure 6.5: Add-on factors for considering CSMA

Let us have a look at the add-on factors that deliver more attractiveness to
adopting CSMA:

Implementation ease:
Accelerated growth of digitalization over the years has led to
rendering traditional security models top-heavy and difficult to
manage.
CSMA is distinctively tailor-made to enable simplicity and
efficiency in designing, deployment, and maintenance.
Practicality:
Old-style cybersecurity policies, methods, and techniques are
complicated due to cloud computing technology, distributed data,
and uncontrolled devices.
The CSMA strategy presents a practical way to deal with
aspects/components that do not lie inside the traditional network
perimeter of an enterprise (business).
Agility:
This has been previously discussed. The very nature of the CSMA
approach renders an enterprise (business) to respond to security
and expansion in a more agile manner.
The benefits of modularity, scalability, and collaboration are as
follows:
Nodes can be modified
(added/removed) relatively easy
A central control point can be used
to monitor and control new
locations

Conclusion
CSMA is a modern approach (conceptual) to build a security architecture to
address the challenges of a distributed enterprise (business) and to say the
least, the continually evolving (changing) threat landscape/attack surface.
Various benefits of adopting CSMA in the wake of embracing digital
transformation and the evolving sophistication of cyber threats are
deployment of security controls where needed the most, establishment of
granular access controls, prioritization of identity-centric security,
collaboration amongst security tools/solutions/services, provision of
centralized management, enablement of end-to-end automation, increased
visibility and control, enhancement of scalability and flexibility,
establishment of a strong framework to strengthen the overall cybersecurity
posture and enhancement of business resilience.
Conclusively, the adoption of CSMA is an involved process and worth the
investment by an enterprise (business). It delivers benefits swiftly in this
progressive journey wherein every step taken leads to improvement in
cybersecurity posture, simplification of operations, and positioning an
enterprise (business) for a brighter future.
In the next chapter, we shall re-examine what necessitates the adoption of
CSMA, the benefits of leveraging CSMA, discuss the characteristics of a
CSMA strategy, delve into a few target use cases, and learn about the
pitfalls of not leveraging CSMA.
Points to remember

Principles for an effective CSMA involve the adoption and


deployment of cybersecurity tools/solutions/services that are not only
fit for purpose but also work within and across the ecosystem
(integrated)
An effective CSMA enables intelligence sharing, automated and
coordinated response to cybersecurity incidents, and simplification of
operations
An effective CSMA transforms an enterprise (business) legacy fit-to-
purpose ecosystem from a standalone/siloed to a fit-to-purpose and
integrated ecosystem

In summary, the adoption of CSMA enables enterprises (businesses) to stay


business (cyber) resilient in this ever-evolving digital world, maintain
customer trust, and reduce the financial impact of security incidents.

Key terms

Open Authorization: OAuth 2.0 is the industry-standard protocol for


authorization.
OpenID Connect: It is an identity layer on top of the OAuth 2.0
framework. It allows third-party applications to verify the identity of
the end-user and to obtain basic user profile information.
System for Cross-domain Identity Management: A specification
and an open standard designed to manage user identity information.
Open Policy Agent: OPA is an open-source engine that provides a
way to write policies as code and then use these policies for decision-
making.

1. https://cybersecurityventures.com/cybercrime-damages-6-trillion-by-
2021/

Join our book’s Discord space


Join the book's Discord Workspace for Latest updates, Offers, Tech
happenings around the world, New Release and Sessions with the Authors:
https://discord.bpbonline.com
CHAPTER 7
Benefits of Adopting Cybersecurity
Mesh Architecture

Introduction
Primarily in Chapter 1, Cybersecurity: A Dynamic Changing Paradigm, we
had a look at the chronology of the evolution of cybersecurity, and some
major trends that had a noteworthy impact. Chapter 2, Cybersecurity:
Understanding Today’s Security Challenges, dealt with understanding
distributed systems (working and challenges) and overall cybersecurity
challenges in the digital age. Whereas, Chapter 3, Emerging Cybersecurity
Trends, covered the emerging cybersecurity trends and the importance of
cyber resilience.
In Chapter 4, The Need for Cybersecurity Mesh Architecture, we first inked
the need for CSMA and the benefits of its adoption. Chapter 5,
Fundamental Components of Cybersecurity Mesh Architecture, covered the
fundamental components of CSMA delivering a framework for security
infrastructure. In Chapter 6, How to Effectively Adopt Cybersecurity Mesh
Architecture, we discussed what enterprises (businesses) can expect when
adopting the architecture and how they can effectively prepare for the
future.
In this chapter, we will re-examine the need for CSMA, the benefits of
adopting the same, and the pitfalls of not leveraging it.

Structure
In this chapter, we will cover the following topics:

What is the necessity of CSMA


Benefits of leveraging CSMA
Characteristics of a CSMA strategy
Target use cases
Features to be considered for CSMA solutions
Pitfalls of not leveraging CSMA

Objectives
We are living in a world that is witnessing an ever-increasing digital
acceleration. This is leading to a swift adoption of new technologies across
the IT ecosystem (including cloud migration). It is not surprising to see
enterprises (businesses) moving first (say even to multi-cloud) and later
asking themselves how to manage and secure these environments.
We also have many organizations with standalone/silos
tools/solutions/services. This leaves their cybersecurity teams to manage the
mesh and complexities of the resulting makeshift arrangements.
These standalone/siloed environments, related complexities, and visible
gaps provide opportunities for cyber attackers to exploit. This is further
exacerbated by resource and skill gaps in cybersecurity (Refer to section
Trend 14: Widening IT skills gap in Chapter 3, Emerging Cybersecurity
Trends).
It is time for enterprises (businesses) to transform their present approaches
to cybersecurity that have a standalone/siloed approach. For this
transformation, they need to adopt CSMA as the baseline (building block).
This would enable them to integrate their distributed ecosystem and create
consistency, interconnectivity, collaboration, and automation across the
various tools/products/services to drive their digital acceleration initiatives.
In the subsequent sections of this chapter, we shall re-look at why CSMA is
necessary to be adopted by enterprises (businesses), cover its benefits,
discuss what a CSMA strategy should look like, take a deeper look at
CSMA target use cases, and examine the downsides of not leveraging
CSMA.

What is the necessity of CSMA


With the acceleration of remote work, adoption of public and private cloud,
IoT intermeshed with traditional IT, the proliferation of ransomware, and
lower barrier entry for startups to challenge large enterprises, there has
never been a more critical need to adopt a CSMA. The network no longer
has an edge, trust is at zero, assets are proliferating and managed by
different teams, and visibility is partitioned to siloed teams.
Without a centralized place to assess if appropriate layers of security
controls are properly implemented across the network, enterprises face a
more complicated challenge in identifying and managing their attack
surface than ever before.
The centralization of security control assessment enables enterprises to save
operational overhead and prioritize business-impactful risk. With a
centralized platform for CSMA, smaller teams achieve more, larger teams
communicate and coordinate more effectively, and siloed teams leverage
one another’s tools to align to risk reduction based on business impact and
not by irrelevant risk scores.

Benefits of leveraging CSMA


The benefits of adopting CSMA have been described in detail in section
Benefits of CSMA, Chapter 4, The Need for Cybersecurity Mesh
Architecture. Presenting in this section is a fresh perspective of describing
the benefits in an abridged manner.
Refer to Figure 7.1 that depicts the benefits of leveraging CSMA:

Figure 7.1: Benefits of leveraging CSMA

Enhanced agility and flexibility


Due to leveraging CSMA for collective consumption, an enterprise’s
(businesses’) security tool/service data integrates consistently and is
structured. This leads to reliable validation and security policy enforcement.
With this, it is possible to deploy new applications and infrastructure
repeatedly and scalable (swiftly). Also, any nonconformities are identified
regularly and submitted for remediation automatically.
As the utility of data sets increases, subsequent increases are seen in the
return on investment (RoI) of existing technologies (security and
infrastructure). Cybersecurity teams are no longer required to do any
manual intervention with this data.
In summary, each cybersecurity solution (monitoring/alerting) in CSMA is
intertwined with one another. This provides a mesh in the ecosystem that
can be navigated easily.

Enhanced security intelligence


Security intelligence (cross-domain) is a reality since data is meshed
together in CSMA. This allows all teams to entrench domain intelligence
for their decision-making purposes.
Such security intelligence, when enabled, enables cybersecurity teams to
make informed decisions that are non-disruptive while responding to
security incidents (aka coordinated response). Since security and
infrastructure services data is accessible for immediate analysis, as a
resultant this provides:

Availability of asset-wide security intelligence


Identification of control gaps
Visibility of network connections and dependencies
Understanding of the business impact of quarantine/changes in the
network
Identification of insider threats

Due to the above-mentioned aspects, cybersecurity teams can leverage


integrated security data to provide more contextual-centric responses and
communicate discoveries (findings) via visualizations in an effective
manner. As a result, analysis is reduced to minutes from hours during
incident response.
The benefits for vulnerability management teams are also immense, as they
can:

Swiftly identify asset owners


Have deeper visibility of business (applications) and network context,
thus can:
Prioritize assets
Understand which vulnerabilities are the most prevalent
Effectively communicate risk

Overall, CSMA leads to the elimination of manual activities such as control


analysis of security gaps and remediation.
Let us consider an example of SIEM. Due to the integration of security
tools/solutions/services (conversion of logs), the data aggregated by SIEM
can be used across cybersecurity teams on an immediate basis for business
risk reduction proactively (since prioritization of most exposed assets is
known contextually). This is also a major consideration when arriving at
ROI.
The key here is the accessibility of security data across cybersecurity teams
that provides them the ability to:

Derive business risk reduction through quantified and qualified


methods
Communicate with business (non-technical staff) effectively

Not only the cybersecurity teams but also the business teams are enabled to
communicate technical subjects with their teams.

Enhanced incident response


Security intelligence (cross-domain) provides maturity to security
processes. By virtue of this security, operations become increasingly
efficient. In the past before leveraging CSMA, cybersecurity incident
response teams would be subjected to correlating huge volumes of data sets
manually to address questions related to asset intelligence. This would lead
to stretching both mean time to resolve (MTTR) and mean time to
containment (MTTC).
Generally, the context of most cybersecurity incidents and alerts is very
limited and needs enrichment on a manual basis. Limited context means
incidents/alerts may contain only:

An IP address
A hostname
An interface
Username (in case of insider threat)

Integration of multiple data sources in case of CSMA lends cybersecurity


analysts to address the following queries swiftly and effectively:

Which device is tied to the IP address (highlighted in the security


incident/alert)?
Which part of the core business function/service is the device a part
of?
Is sensitive data stored in the device?
Will there be any business downtime due to containment activities?
Which devices have an affected user logged into?

This exemplifies the need for CSMA. The limiting factor with previous
ecosystems/architectures was the non-interoperability of cybersecurity
tools. This inadequacy hampered swift decision-making for effective
incident management.
In summary, CSMA brings about interoperability and cohesiveness of
cybersecurity tools/platforms, resulting in effective decision-making owing
to enhanced visibility of contextual information and a reduction in the
MTTR of cybersecurity incidents and responses.

Efficient compliance
Leveraging CSMA leads to a reduction in compliance-related obligations
and risks. This is due to the simplification of compliance monitoring, where
the assessment process is centralized through maximization automation.
In addition, CSMA assists security teams in comparing their enterprises
(businesses) standards to industry standards, viz., the National Institute of
Standards and Technology (NIST).

Enhanced savings
The cost of ransomware attacks and cybercrime attacks is increasing year
on year. Digital transformation incurs its own cost, too, especially if an
enterprise’s (business) infrastructure/architecture has to be redesigned.
However, the reduction in costly attacks and business downtime realized by
leveraging CSMA far outweighs any initial cost.
CSMA saves precious time and resources for cybersecurity teams by
eliminating the need for aggregating data manually and the use of multiple
query languages.

Characteristics of a CSMA strategy


The main characteristics of CSMA include:

Integration
Comprehensiveness
Automation

Refer to Chapter 6, How to Effectively Adopt Cybersecurity Mesh


Architecture, section, Key aspects for CSMA adoption, for more details.
These characteristics are vital for the reduction of complexity and increase
of overall security effectiveness. To achieve this, enterprises (businesses)
need to integrate their CSMA strategy across various cybersecurity
tools/solutions/services portfolios.
Consequently, enterprises (businesses) can utilize this comprehensive
portfolio of interconnected tools/solutions/services to solve their
cybersecurity challenges across key components of CSMA. Refer to
Chapter 5, Fundamental Components of Cybersecurity Mesh Architecture,
section, The key components of CSMA for details.
It is all the more important for enterprises (businesses) to break down
vendor silos and standalone technology. This can be enabled and supported
through a broad open ecosystem of technology partners, which allows
flexibility in deployments. Also, such an ecosystem then consolidates
security operations and visibility across. Not only does this setup lead to the
implementation of an integrated and automated security experience, but it
also preserves the existing tools/solutions investment.
Enterprises (businesses) can gain the following benefits of leveraging a
CSMA today:

Obtain visibility (in-depth) across its ecosystem.


Manage and deploy solutions centrally (leading to consistent
configurations and policies).
Leverage cybersecurity intelligence (cross-domain) to obtain
protection on a near real-time basis for cyberattacks.
Automate actionable responses across the ecosystem.
Data normalization and correlation.
Employ advanced data normalization and correlation techniques to
ensure consistency and accuracy in the information retrieved from
different security tools. This process aligns disparate data formats,
standardizes data representations, and identifies relationships between
data points, enabling comprehensive analysis and contextual insights.

It is again vital to re-emphasize the importance of deployment of data


normalization and correlation practices in the context of CSMA. These
practices ensure consistent and accurate retrieval of information from
various cybersecurity tools/solutions/services.

Target use cases


Refer to Figure 7.2, which lists some of the target use cases of leveraging
CSMA:
Figure 7.2: Target use cases of leveraging CSMA

Security operations
CSMA provides cybersecurity teams with an integrated (comprehensive)
view of the security stack, thereby meaningfully enhancing security
operations.
Also, this enables the cybersecurity teams to swiftly and conveniently
obtain deep insights, recognize potential threats, and undertake proactive
steps to mitigate risks.

Vulnerability management
There has been a multi-fold increase in phishing incidents over the years,
and ransomware attacks occur every other hour. Current IT systems are,
therefore, very vulnerable. Not only is it the existing vulnerabilities, but
also there is the threat of a zero-day attack, an attack by a method
previously unknown.
CSMA enables the identification and analysis of vulnerabilities across the
security stack thereby lending support to vulnerability management.
It helps cybersecurity teams prioritize and address vulnerabilities basis their
business impact (contextual) and associated risk levels.
CSMA approach also prepares an enterprise’s (businesses’) security
infrastructure for zero-day attacks.

Compliance management
CSMA provides a centralized platform for compliance assessment across
various cybersecurity tools/solutions/services thereby simplifying
monitoring of compliance.
This eventually enables effective tracking of compliance status,
identification of gaps, and streamlining the reporting process.

Features to be considered for CSMA solutions


CSMA vendors/players must provide a platform that gives security
practitioners, executives, and security teams swift and easy access to
contextual insights and responses across the security stack.
This is possible by building a comprehensive graph of APIs (covering cloud
and on-premises security tools). The platform should enable users to
navigate and understand the relationships between different data points
effectively.

API Graph
It will be essential for a CSMA vendor/player to establish an interconnected
(dynamic) graph of the APIs exposed by various security tools (covering
cloud and on-premises-based solutions).
This holistic graph will provide a comprehensive view of the security stack
and will assist in understanding how the data exposed by each tool relates to
others.

Data normalization and correlation


A CSMA vendor/player will be required to employ advanced data
normalization and correlation techniques. This will be required to make
sure that the information retrieved from different security tools is consistent
and accurate.
This will lead to an alignment of disparate data formats, standardization of
data representations, and identification of relationships between data points.
Eventually, this will lead to the enablement of a comprehensive analysis
and contextual insights.

Natural language processing


A layer of natural language processing (NLP) should be integrated into
the CSMA services. This will allow users to interact with the platform using
natural language queries (thereby leveraging cutting-edge NLP algorithms).
The CSMA vendor/player should be in a position to understand the intent
behind the queries and extract relevant information from the security stack,
providing precise and contextual responses.

Generative AI for context enhancement


The CSMA vendor/player should consider utilizing generative AI
algorithms to enhance contextual understanding and provide more accurate
and insightful responses (through leverage of generative models, such as
language models (LMs).
The vendor/player will thereby be able to generate free text responses that
will go beyond simple keyword matching. This will enable users to gain a
deeper understanding of the security stack and its implications.
Overall, a CSMA player/vendor should work towards a transformational
platform by taking leverage of an API graph, data normalization, and
correlation, and advanced NLP techniques, covering generative AI. Such a
platform will empower security practitioners, executives, and security
teams with contextual responses and insights.

Pitfalls of not leveraging CSMA


In this section, we will analyze the reasons for adopting CSMA, its key
components, and the benefits of leveraging the same. Let us look at the
other side of the coin and seek the pitfalls of not leveraging CSMA.
Refer to Figure 7.3, which lists the major pitfalls of not leveraging CSMA.
Figure 7.3: Pitfalls of not leveraging CSMA

Increased security overheads


So far, we have realized that one of the biggest challenges for enterprises
(businesses) is siloed data/expertise. When ecosystems do not have
anything in place for aggregating each of these silos into a central repertoire
(such as a data lake or SIEM), they require more staff to analyze
cybersecurity events and threats.
In the absence of CSMA, cybersecurity teams oscillate between databases
and consoles, use each tool’s query language (which is proprietary to the
tool), and manually compile and analyze the data of each asset. All this is to
obtain an understanding of its connections and validate security control
implementation.
Cybersecurity teams, instead of involving themselves in value-added
business risk reduction activities, spend most of their valuable time in
simple auditing activities (which is a monotonous, tedious, and intricate
process).
Let us consider the following real-life situation related to “increased
security overheads”:

Problem statement: An enterprise (business) is required to retrieve


data from two platforms: the vulnerability management console and
the other is the asset inventory management system. Identification of
gaps between inventories and assets that are not monitored requires
data sets to be merged and then analyzed.
Every other data point, such as the EDR agent, adds a
data layer that is required to be retrieved and analyzed.
Moreover, every asset that is not monitored will require
a remediation ticket (creation and submission).
Challenges: Simple activities such as this create a momentous
challenge as and when the sheer volume of security control gaps
surpasses the ability to remediate the gaps.
In the absence of practical implementation of
automation for identifying gaps in security controls,
enterprises (businesses) will continue the practice of
expanding headcount and reassigning their valuable
analysts to low-value-adding manual activities.
Solution
Prioritize analysis
If the activity is repetitive, automate it
Aggregate data sets contextually
Conduct predefined searches across
data sets automatically
Manage security control gaps that
are identified through appropriate
triggers
Benefit: Data being centralized and accessible leads to gaining
intelligence about assets. This enables cybersecurity teams to reduce
MTTR and MTTC of security incidents.
Increased security risk
Let us consider the following real-life situation related to increased security
risk:

Problem statement: As enterprises (businesses) expanded beyond the


four walls of the office, cyber assets moved out of the perimeter to
potentially become an ingress point (relatively easy to exploit) for a
breach.
Understanding the attack surface requires a repository
(central) of all assets in the ecosystem, software running
on those assets, vulnerabilities of each asset, and the
way they communicate with each other. The absence of
this makes the process of understanding the attack
surface time time-consuming and often difficult to
achieve.
Challenge: Due to siloed network/security tools/platforms, the
approach to risk becomes vacuumed. It is like missing the woods for
the trees for network/security teams.
In the absence of data convergence, a huge effort is
required to manually correlate across huge data sets to
be able to understand the attack surface. Moreover, any
manual analysis is as good at a point in time and could
quickly become irrelevant.
Solution: Leverage CSMA to identify and automatically enable the
mapping of the attack surface by converging identity, application,
network, asset data, and security. This also leads to saving substantial
effort by cybersecurity teams.
Benefit: The availability of converged data leads cybersecurity teams
to better understand each other’s viewpoints on risk, such as unusual
user behavior or network access related to exploitation.

Inefficient security incident management


Let us consider the following real-life situation related to inefficient security
incident management:
Problem statement: The availability of limited business context in
decision-making, results in extended MTTR. There are risks of
unexpected business disruptions when the implications of
quarantine/service suspension are not well understood.
We often observe that inadequate awareness of business
context often thwarts the ability of cybersecurity
operation teams to respond due to fear of impacting the
business.
Challenge
Increase in dwell time of attacks
Increase in the probability of ransomware attacks, and
Increase in disruptions (unpredictable) across enterprise
environments
Solution: Leverage CSMA to automate security data aggregation and
analysis and enable security intelligence (cross-domain) to provide an
informed view of risks and threats.
Benefit: Cybersecurity teams can reduce MTTR in situations where
they can:
Identify the owner of a vulnerable server on an immediate basis
Identify the users of the server based on login activity
Identify the connections of the asset to other assets
Identify security events impacting the server

Incident responses can be adjusted and refined when dependencies (asset


and application) are understood, minimizing business disruptions.

Restricted decision-making
Let us consider the following real-life situation related to restricted
decision-making:
Problem statement: Siloed ecosystems depend on data sets from their
tools (individual and not integrated), thus decisions are arrived at
without business context.
Enterprises (businesses) have assets that run into
hundreds of thousands with millions of vulnerabilities.
In this situation, it is not possible to resolve all risks. To
manage remediation, enterprises (businesses) either
depend on vendor-provided risk scores or any other
form of aggregated score.
Challenge: Using risk scores that lack business (and network) context
leads to ineffective risk prioritization and incorrect business impact.
Without including context across the business and network, pursuing a
non-contextually relevant risk score (lacking business/asset
intelligence) results in ineffective risk prioritization, risk reduction,
and uncertain business impact.
The absence of business/network context has negative
impacts on different teams:
Vulnerability management teams make incorrect decisions related
to the priority of assets and patches
Security operations teams decide basis the context of the
technology stack from SOC rather than the business
Networking teams decide based on routing and ambiguous IP
addresses
Solution: Leverage CSMA to enable domain intelligence (cross-
domain). This aids the convergence of siloed knowledge and data
leading to augmenting the cybersecurity team’s decision-making
ability.
Benefits: A meshed ecosystem covering network and security services
helps vulnerability management teams swiftly identify the vulnerable
assets that are a part of crown jewels applications. This kind of
analysis will otherwise require domain expertise across networking,
vulnerability analysis, and penetration testing which is not easy.
By leveraging CSMA, the data aggregation, and analysis is automated
leading to abstracting the result into a visual and interactive format that
simplifies the complex.
The real benefit of adopting CSMA is to:

Make security accessible to all


Overcome Silos; and
Enable well-informed vulnerability management decisions

Conclusion
CSMA works towards addressing the challenges presented by the
distributed enterprise (business) and the continuously changing threat
landscape/attack surface.
CSMA’s adoption leads to deployment of security controls where they are
needed the most, establishment of granular access controls, enhanced
facilitation among security tools/solutions/services, increased visibility and
control (centralized) across the ecosystem, increased agility due to
reduction in deployment times and acceleration of digital transformation,
increased resiliency by way of understanding interdependencies (context
cross-domain) leading to better uptime and recovery, increased efficiency
by allocating more experienced staff to higher-value activities and reduced
risk by implementing business-centric risk reduction programs.
Overall, CSMA presents a robust framework that enables the strengthening
of an enterprise (business) cybersecurity and protecting their valuable
assets.
In the next chapter, we shall compare CSMA with traditional defense-in-
depth approach, re-visit the salient points and goals for CSMA, present a
list a systematic approach to implementing CSMA, list some of the key
performance indicators) for assessing the effectiveness of the
implementation of CSMA and introduce the ten commandments of CSMA.

Points to remember
CSMA is a baseline foundation that enables disparate security
tools/solutions/services to work in an integrated manner to create a
dynamic security ecosystem.
As enterprises (businesses) embrace digital technologies, CSMA lends
a scalable and flexible base that enables bolt-on security for assets in
hybrid and multi-cloud environments.
Adoption of CSMA provides major benefits to enterprises
(businesses). It leads to:
Reduction (significant) in the manual analysis of security data
(freeing up cybersecurity teams for more value-added activities)
Consolidation of security analytics (predictive)
Consolidation of cybersecurity posture management
Enhanced security incident response
Effective business risk reduction

Key terms

Mean time to containment: Average time it takes for an enterprise


(business) to deal with a security breach or incident post-detection.
National Institute of Standards and Technology: This is an agency
of the US Department of Commerce, which has a mission to promote
American innovation and industrial competitiveness. NIST has
published a cybersecurity framework (CSF). This framework
provides a set of guidelines for mitigating enterprise-wide (business-
wide) cybersecurity risks.
Natural language processing: An interdisciplinary subfield of
computer science and artificial intelligence.
Dwell time: Amount of time a cyber attacker has access to a
compromised system before it is detected.
Return on investment: Approximate measure of the profitability of an
investment.
CHAPTER 8
CSMA Best Practices

Introduction
We have covered various topics in the first few chapters ranging from the
chronology of the evolution of cybersecurity, understanding the distributed
systems (working and challenges), cybersecurity challenges in the digital
age, the emerging cybersecurity trends, and the importance of cyber
resilience.
In the subsequent chapters, we took a first-hand view of the need for
CSMA, and the benefits of its adoption, looked into the fundamental
components of CSMA, and further discussed what enterprises (businesses)
can expect when adopting the architecture and how they can effectively
prepare for the future.
In the preceding chapter, we understood the need for CSMA, its benefits,
and the pitfalls of not adopting it.
In this chapter, we will concentrate on the best practices that will help
enterprises (businesses) to traverse the path to incorporating CSMA. It
brings about an architectural and philosophical change that is an asset to
enterprises (businesses) for all practical purposes.

Structure
In this chapter, we will cover the following topics:

CSMA vs. defense-in-depth approach


Salient points and goals for CSMA
Systematic approach to implementing CSMA
CSMA key performance indicators
The ten commandments of CSMA
Challenges in implementing CSMA

Objectives
In chapters 5-7, we learned that CSMA enables enterprises (businesses)
with the strategy for proactive cyber risk mitigation. The cornerstones of
this approach are aspects of interoperability and collaboration among
cybersecurity teams and tools/solutions. This further nurtured an
interconnected and context-aware cybersecurity ecosystem, which reduces
business risk and enhances operational efficiency by disrupting traditional
cybersecurity paradigms.
We will further discuss in detail how CSMA takes a vendor-agnostic
approach and enables enterprises (businesses) to utilize the strengths of
multiple vendors instead of living with the constraints of a single vendor’s
ecosystem.
This creates flexible (scalable) and adaptable enterprises (businesses). They
can select the most appropriate security tools/solutions/products and
technologies for their specific needs and stay ahead of emerging threats.
In this chapter, we shall revisit the goals of CSMA, seek a step-by-step
approach to implementing it, and list the KPIs for assessing the
effectiveness of the implementation.

CSMA vs. defense-in-depth approach


Over the years, enterprises (businesses) have relied on a defense-in-depth
strategy for cybersecurity. Owing to this, they have implemented various
security tools/solutions to address specific needs, leading to these
enterprises (businesses) facing the challenge of security tool/solution
sprawl.
Moreover, this approach has led to a complex and fragmented security
infrastructure. Disjointed tools/solutions have become a scary proposition
for enterprises (businesses) and resulted in an increase in operational costs
and a decrease in efficiency. Management and integration of this situation
require a sweeping transformation.
Overall, a cybersecurity mesh architecture:
Advocates a security framework that is integrated and interconnected,
thereby providing a solution to the challenge of security tool/solution
sprawl.
Promotes interoperability and collaboration among security
tools/solutions.
Does not rely on standalone/siloed security tools/solutions/services.
Leads to reducing the complexity of managing disparate tools.
Integrates disparate security tools/solutions into a unified ecosystem
(by eliminating redundant/overlapping security tools/solutions).
Leads to streamlining security operations (efficiency) and saving costs.
Enables seamless communication and sharing of data among security
tools/solutions, thus leveraging their collective intelligence.
Facilitates a well-coordinated defense strategy that is rich in context.
Leads to enhancing the overall effectiveness of the security
infrastructure (by streamlining an approach to protecting digital
assets).

Let us now examine (refer to Table 8.1) the major differences between
CSMA and the traditional defense-in-depth approach:

Framework CSMA Defense-in-depth

Approach Integrated Layered


Interconnected Multiple defenses

Focal point Centered on identity Centered on network/site

Communication Seamless among key components Limited communication


(Refer to Chapter 5, section, The key components of between layers
CSMA)

Topology Centralized Distributed

Contextual Strong context awareness due to continuous Limited awareness of


awareness analysis and monitoring context
Scalability Scalable Complex and difficult to
scale

Agility Adaptable to emerging technology Limited adaptability to


emerging technology

Collaboration Promotes collaboration and integration Limited collaboration


among vendors

Operations Streamlined security operations due to Tool/solution sprawl


consolidation leading to inefficiency

Defense Proactive Reactive


strategy Enhanced business context leading to effective Siloed approach due to
threat prevention, detection, and response fragmentation

Table 8.1: Comparison of CSMA vs. defense-in-depth approach

Salient points and goals of CSMA


In Chapter 5, Fundamental Components of Cybersecurity Mesh
Architecture, we discussed the key components of CSMA and the outcome
of its adoption. In Chapter 6, How to Effectively Adopt Cybersecurity Mesh
Architecture, we examined the importance of integration,
comprehensiveness, and automation in CSMA and the key factors to
consider for “its adoption. Furthermore, in Chapter 7, Benefits of Adopting
Cybersecurity Mesh Architecture, we covered the characteristics of a
CSMA strategy.
Figure 8.1 provides a view of the salient points and goals of CSMA:
Figure 8.1: Salient points and goals of CSMA

In this section, let us re-examine the salient points and goals of CSMA:

Promotes security centered around identity


Refer to Chapter 5, section Does CSMA work with zero trust,
wherein we mentioned that zero trust is a security model that is
centered around identity and has a strong focus on user
authentication and authorization.
Salient point/goal of CSMA: Zero trust adoption is streamlined
through ease of transition to identity-centric security.
Improves integration of security ecosystem
Many enterprises (businesses) have a complex range of disparate,
disconnected security tools/solutions.
Salient point/goal of CSMA: Bring about a reduction of
complexity and improvement in performance by accelerating
security integration and collaboration.
Refer to Chapters 6 and 7 for more details regarding
collaboration and integration.
Improves interoperability
Generally, enterprises (businesses) face challenges due to the
absence of interoperability among standalone security
tools/solutions from different vendors, which leads to gaps.
Salient point/goal of CSMA
Bridge these gaps through defining
a framework for cooperation and
collaboration.
Supports growth by using plug-in
APIs to more easily support
extensions, customization,
analytics, and support for new
regulations and standards.
Refer to Chapter 7, section Benefits of leveraging
CSMA, for more details regarding interoperability.
Simplifies security design
Salient point/goal of CSMA: The key components enable
enterprises (businesses) to plug in security tools/solutions as
required to meet security requirements in a structured manner.
Refer to Chapter 5, section, The key components of
CSMA, and Chapter 6, section Build the key
components, for more details.
After considering the above, we should discuss a vital
point: the approach to implementing CSMA should be
vendor-agnostic. For details on this subject, refer to
Chapter 6, section, Build the key components.

CSMA needs to be vendor-agnostic


Let us assess this subject through a real-life scenario detailed as follows:

Problem statement:
In today’s cybersecurity field, enterprises (businesses) rely on
multiple vendors for security tools/solutions.
A unique set of expertise and capabilities is offered by each
vendor, and enterprises (businesses) generally select different
products from different vendors to address their specific security
requirements.
Challenge:
Such an approach leads to challenges such as vendor lock-in and
interoperability among security tools/solutions.
Solution:
CSMA, by default, builds in a vendor-agnostic approach that aids
in designing a resilient and scalable (flexible) security
architecture.
Due to this approach, enterprises (businesses) can utilize the
strengths of multiple vendors and avoid the limitations of the
ecosystem of a single vendor.
Benefits:
Enterprises (businesses) can choose the most befitting security
tools/solutions/products basis specific needs instead of restricting
themselves to a specific vendor’s offerings.
Vendor-agnostic approach supports and provides a fillip to
integration and collaboration among vendors.
Enterprises (businesses) can utilize the expertise and collective
intelligence of the cybersecurity ecosystem, due to security
tools/solutions/products from different vendors that can
seamlessly communicate and share information.
Collaboration leads to the enhancement of cyber defense
capabilities and enables enterprises (businesses) to effectively
respond to cyber threats.
Vendor-agnostic approach also provides adaptability and
flexibility.
enterprises (businesses) can incorporate new and emerging
solutions and technology into their cybersecurity ecosystems, as
they no longer are limited by a single vendor’s roadmap.
Overall, this approach enables enterprises (businesses) to steer
ahead of evolving cyber threats by leveraging the latest
innovations in cybersecurity.

Systematic approach to implementing CSMA


In continuation with Chapter 6, section How to get started with CSMA, let
us re-visit the steps for systematically implementing CSMA. This will aid
as a quick recap.
Figure 8.2 lists a step-by-step approach to implementing CSMA:

Figure 8.2: Approach to implementing CSMA

1. Assess:
1. Assessing the present state and requirements of the enterprises’
(businesses’) technology environment and cybersecurity posture
2. Identifying the gaps and opportunities for improvement (OFI)
3. The key steps are:
1. Conducting a security audit, risk assessment, and maturity
assessment
2. Defining the security goals and objectives

2. Design:
1. Designing the CSMA framework
2. Selecting the cybersecurity tools/solutions and vendors to be
included in the CSMA ecosystem
3. The key steps are:
1. Defining the cybersecurity requirements and specifications
2. Evaluating and comparing the security tools/solutions and
vendors
3. Creating the cybersecurity roadmap

3. Deploy:
1. Deploying and configuring the cybersecurity tools/solutions
2. Integrating them with the CSMA key components (Refer to
Chapter 5, section The key components of CSMA and Chapter 6,
section The key components of CSMA for details)
3. The key steps are:
1. Testing the cybersecurity tools/solutions
2. Establishing the security policies and rules
3. Enabling the security data and communication channels

4. Operate:
1. Operating and monitoring the cybersecurity tools/solutions and
the CSMA ecosystem (ensure proper and effective functioning)
2. The key steps are:
1. Managing and maintaining the cybersecurity tools/solutions
2. Collecting and analyzing the cybersecurity data
3. Responding to security incidents and alerts
5. Optimize:
1. Optimizing and improving the cybersecurity tools/solutions and
the CSMA ecosystem (ensuring alignment and updation with the
changing industry trends).
2. The key steps are:
1. Reviewing and evaluating the performance and outcomes
2. Identifying and implementing the cybersecurity best
practices and enhancements
3. Planning and executing the cybersecurity changes and
upgrades

CSMA standouts (in comparison to other architectures)


Refer to Chapter 5, section CSMA’s scalability relative to other
architectures, where this subject was first discussed. Let us summarize the
ways CSMA is different from other security architectures:

Way 1: Composability
Cybersecurity controls and functions are modular and can be
deployed per the needs/preferences of each asset/user
Not constrained by a single vendor/platform
Way 2: Interoperability
Cybersecurity tools/solutions can communicate and collaborate
via key components such as data integration, comprehensive
threat intelligence (and analytics), distributed identity fabric,
automated and centralized operations, and consolidated
dashboards (also consolidated policy, posture management, and
modular security controls).
Way 3: Collaboration
Cybersecurity tools/solutions can leverage and share the data and
capabilities of each other
Provide a more consistent and comprehensive cybersecurity
posture

Overall, CSMA:

Enables the communication and coordination between the


cybersecurity teams and stakeholders.
Provides a common language/platform for cybersecurity management
and governance.
Well-suited for Enterprises (Businesses) that need to secure
hybrid/multi-cloud/remote environments (also where resources/users
are diverse and distributed).
Enables a more flexible, scalable, resilient, and collaborative
cybersecurity ecosystem.

CSMA key performance indicators


Enterprises (businesses) need to have KPIs that showcase how the CSMA is
delivering the desired outcome. It is prudent to determine which KPIs are
vital to be tracked and reported.
A high-level KP can be as follows:

How is the CSMA strategy affecting the overall business outcomes?

Alternatively, KPIs can be technical/actionable and include (but are not


limited to):
Figure 8.3 provides a comprehensive list of such CSMA KPIs:
Figure 8.3: CSMA KPIs illustrative list

The above list is only indicative and should be considered as a starter. The
KPIs chosen by an enterprise (business) to be tracked will depend on its
environment (that is unique to that enterprise) and its CSMA ecosystem.

The ten commandments of CSMA


Cybersecurity mesh is an architectural and philosophical change that will be
an asset to your business for many years to come.
Here are ten easily digestible bits of information that will help you get on
the road to incorporating cybersecurity mesh.
Refer to Figure 8.4 that lists the first five commandments of CSMA:
Figure 8.4: CSMA 1-5 commandments

Inventorize – present technology


Understand how the present security system operates (aids in
integration)
Identify weaknesses, inadequacy of tools, and possible points of
failure
Prepare for the transition into a CSMA
Prioritize – risks
Assess risks and potential vulnerabilities
Assess possible issues with the adoption of new technologies
Think like an attacker before implementing new security
tools/solutions
Prioritize the identified vulnerabilities
Security operational teams should identify the most vital security
weak areas that need to be addressed (this aids the direction for
new technology adoption)
Consider enterprise-wide and managerial challenges that may be
affecting security performance (this is beyond the technical
challenges)
Assess security teams’ structure
Ensure clear lines of communication between teams
Implement – best practices
Best practices are a vital aspect of any cybersecurity ecosystem
Follow guidelines, viz., the principle of least privilege, and
integrate any tool/solution that can be integrated (essential to fully
secure resources)
Best practices are likely to change over time, hence it is important
to keep reassessing the policies on a regular basis
Identify – silos
Silos lead to locking meaningful data in separate containers
thereby damaging the overall readiness of cybersecurity
ecosystem
Due to unclear visibility and inadequate accessibility of the
cybersecurity ecosystem, cybersecurity teams are unable to
access/control their tools/solutions intelligently or set policies
proactively
Key to breaking Silos is to move security controls closer to
systems/assets they govern (this leads to decreasing detection and
response times)
Centralize – threat intelligence (analytics)
As discussed above, integrated (unified) cybersecurity systems
not only lead to lowering detection and response times), but also
lead to effective gathering of intelligence and analytics
Centralization of cybersecurity system information
Enables speeding up of security
incident response plans
Leads to a deeper and contextual
understanding of the cybersecurity
landscape

Refer to Figure 8.5, that lists the next five commandments of CSMA:
Figure 8.5: CSMA 6-10 commandments

Leverage – AI and ML
Leverage recent developments in AI and ML for security-related
applications
Vet AI and ML tool providers and ensure they provide holistic
and effective tools/solutions/products
Automate – security stack
Streamline cybersecurity operations through automation
(integrated cybersecurity ecosystem). Automate as many tasks as
possible such as device updates, tasks related to analysis (saves
the crucial time of security staff to spend on evaluation tasks).
Automate threat response and threat analysis to generate swift
security intelligence.
Define security policies and design playbooks.
Identify – skill gaps
It is increasingly hard to find cybersecurity talent.
Different skills are required to manage modern cybersecurity
tools/solutions such as for endpoints, edge devices, cloud assets
and applications, and the skill gap is continuously widening.
Use of CSMA leads to simplification of cybersecurity processes.
It also enables the narrowing of skill deficiencies by assigning
new security tasks to more familiar zones.
External raining or outsourcing skills can be utilized to manage
the more serious skill gaps.
Partner – with ecosystem players
There is no Enterprise (Business) that has all the expertise
required to maximize its cybersecurity posture.
It is vital to contact industry experts regarding security concerns,
potential attack surface vulnerabilities, etc.
Leverage the know-how of the industry experts who are
specialized professionals (better equipped than an average
security team in an enterprise (business).
Improve continuously
Every cyberattack presents an opportunity to:
Evaluate the current policies and
playbooks
Identify areas of improvement for
the cybersecurity systems
Keep pace with the latest industry practices to ensure that the
threat readiness is at par with current potential cyber threats.
Conduct assessments of the security architecture on a regular
basis.

Challenges in implementing CSMA


Enterprises (businesses) may face challenges while implementing CSMA.
The most common challenges are:

Managing the transition from the existing security framework to


CSMA
Managing integration of disparate security tools/solutions into a
unified ecosystem

Solution
Carrying out a thorough assessment of the enterprises’ (businesses’)
cybersecurity infrastructure
Finding redundancies
Prioritizing integration efforts
Designing a roadmap
Collaborating with vendors
Carrying out holistic user awareness training and education to
cybersecurity teams on CSMA, its key components, and benefits
(including the best practices)

Conclusion
Enterprises (businesses) need to adopt CSMA and build a solid baseline for
safeguarding their digital assets. This becomes increasingly important in an
ever-changing cybersecurity threat landscape.
We have seen that the implementation of CSMA presents its challenges.
These can be managed via thorough assessments, vendor collaboration, and
comprehensive user training (education).
In the last chapter, we shall examine use cases in three distinct
environments where CSMA works (viz., work from home, cloud, and
operational technology (OT)) and have a sneak preview of the CSMA
market overview, its growth factors, dynamics, and opportunities for
growth.

Points to remember

CSMA is a paradigm shift in cybersecurity and offers a holistic and


integrated approach to protecting digital assets. Enterprises
(businesses) that adopt CSMA can augment their cybersecurity
posture. By establishing cybersecurity centered around identity, they
can detect and respond effectively to security alerts/incidents.
The key to benefiting from the full potential of CSMA is collaboration
and integration (interoperability).
The tips for an effective implementation of CSMA are as follows:
Carry out a present state security assessment
Understand existing strengths and weaknesses (part of SWOT
analysis)
Adopt a collaborative culture among cybersecurity teams and
vendors (enable seamless communication)
Prioritize the implementation of IAM solutions (identity-centric
security)
Leverage AI and ML for:
Continuous monitoring
Proactive threat detection
Choose tools/solutions that are vendor-agnostic thus providing
flexibility (scalability) and integration
Impart awareness training and education to cybersecurity teams on
CSMA
Carry out phased implementation
Assess and improve cybersecurity posture on a continuous basis

Key terms

Key performance indicators: A measure (quantifiable) of


performance over time for a specific objective.
Opportunities for improvement: In simple terms, it is an opportunity
to improve your enterprise (business) system.
CHAPTER 9
Potential Outlook for CSMA Adoption

Introduction
In the previous chapters, we concentrated on the benefits of adopting
CSMA and explored its best practices. We conducted research on many
aspects of CSMA viz, its necessity, its target use cases, downsides of not
leveraging CSMA, characteristics of a CSMA strategy, its salient points and
goals, key performance indicators (KPIs), and eventually its ten
commandments.
In this chapter, we shall examine some distinct use cases where CSMA
works. We will take a specific example of the healthcare industry and
examine how healthcare systems can leverage CSMA (potentially adopted).
Lastly, we will understand another important aspect of the global market
overview and dynamics of CSMA.

Structure
In this chapter, we will cover the following topics:

Work-from-home environment
Cloud security environment
OT security environment
CSMA in the healthcare system
CSMA

Objectives
We have seen that CSMA is a holistic approach to safeguarding information
assets and infrastructure. Consequently, CSA works in varied situations,
and this chapter aims to examine different use cases, viz., work-from-home
environments, cloud environments, and operational technologies (OT)
environments. These environments present variations/distinct security
challenges, and CSMA can address these by significantly contributing to
maturing security in each of these scenarios.

Work-from-home environment
COVID-19 brought about a transformational change wherein employees
started to work from home on devices that enterprises (businesses) had
no/little control over. We say transformational for the reason that these
enterprises (businesses) were pushed into a rare situation that required
fundamentally new ways of working.
CSMA approach for security work-from-home environment entails:

Deployment of zero-trust network access (ZTNA)


Usage of endpoint detection and response (EDR), and
Implementation of best practices for home network security

All of the above-mentioned key factors will be discussed subsequently.

Zero-trust network access


Problem statement
Remote workers have since ages been using virtual private networks
(VPN) to access enterprise (business) networks. By using a VPN,
enterprises (businesses) can limit access to their infrastructure and systems
based on the user’s network connection. It was presumed that if a user is on
a secure network, he/she can be granted access to the systems.
This reasoning will hold if there is strong physical security guarding access
to buildings that house your systems and only people in those buildings
have access to those systems. As soon as you allow a remote worker to
access those systems, this security control does not hold anymore.
Logically networks can be extended to remote locations by using VPNs.
Communications are encrypted (between endpoints and the central
network). This is to thwart others on the network from understanding what
is being communicated. Unfortunately, this level of security is no longer
adequate.
Perceived risk
An employee (user) can always have a reason (legitimate) to use a VPN to
access an enterprise (business) resource however the user may not have any
reason to access many other resources. Consider a scenario where a laptop
that was configured for remote access is lost/stolen. This scenario to the
enterprise (business) could seem like a legitimate use case of the VPN
whereas, in reality, the lost/stolen laptop (endpoint) is being used to steal
confidential data from the network.
Solution
The need of the hour is for a better approach to address these scenarios. The
approach is to follow a zero-trust network access (ZTNA) approach. In
this approach, an enterprise (business) will not trust a user simply because
they have access to a VPN. The fundamental difference in this revised
approach is that not only the user but also the device must be authenticated.
With this approach, an attacker would not be able to gain access to
enterprise (business) infrastructure just by gaining access to a VPN but
would need to possess the device that was allowed access to the
infrastructure (that will require authentication and authorization checks).

Endpoint detection and response


Problem statement
Devices used for remote work can be used from home, cafes, and flights,
from anywhere where there is availability of wireless connection. A
publicly accessible network can be used to connect to an office
environment.
Perceived risk
With this kind of setup, cyber attackers can explore ways to use these
networks to perpetrate attacks on other devices on the network. This can
also be used to spoof users into connecting to a network that seems
legitimate but is controlled by attackers.
Solution
CSMA ensures that all endpoints in the enterprise (business) are configured
and monitored to detect and prevent attacks. It is vital to maintain endpoint
protection in the wake of a fast and constantly changing threat landscape. A
holistic endpoint detection and response services are important to block and
remediate attacks (e.g., ransomware attacks) on the endpoints.

Home network security


Problem statement
It is quite reasonable to assume that a normal user of a home network may
be running an insecure network. Moreover, home networks are typically
secured using wireless routers that are shared among other users.
Perceived risk
Home wireless routers may be configured with weak administrator
passwords or they may have older software with known vulnerabilities that
could be exploited by an attacker.
Solution
CSMA can lend a helping hand to extend enterprise (business) security to
networking devices in the home. This not only provides visibility into the
state of home networks that are used for work but also control over the
extended enterprise (business) network.
Cloud security environment
Enterprises (businesses) are swiftly adapting cloud computing services
because they offer major advantages. New technologies such as cloud
computing services are associated with new ways of doing things (aka
developing and managing software) and also present distinct security
challenges. To understand the specific security challenges of cloud
computing, let us review some characteristics of cloud-native applications.
Problem statement
Characteristics of cloud-native applications
The deployment model or applications that run in the cloud are markedly
different from applications that have traditionally been used in on-premise
systems. In the case of on-premises applications, physical servers and
virtual machines (VMs) are used widely, however, cloud platforms use
containers for optimization. The distinct advantage of using containers is
that it uses fewer resources as compared to VMs. Hence in the case of
containers, a single server can run more applications effectively as
compared to VMs.
Management challenges grow with the growing number of containers. The
challenge lies in the deployment of containers such that servers can
effectively be used, as well as in securing the containers. Containers are
required to be monitored regularly as well to ensure they are running as per
expectations. Upon failure of a container, an application or associated
service will become unavailable until a replacement container is brought up.
These and other operational challenges encountered during large-scale
deployment of containers are addressed by Kubernetes. Kubernetes is a
platform for orchestrating the dynamic deployment of containers on a large-
scale basis. This is useful when applications are being developed using
architectures based on microservices.
In microservices, complex systems are disintegrated into small functional
units/services that independently operate. Every microservice can be
deployed in its container, allowing administrators to update and manage
microservices without the fear of disruption of other services.
Perceived risk
Cloud computing has associated security and operational issues that need to
be addressed. For example, the OWASP Top 10 list recognizes some
important security issues for web applications that need to be managed.
Another aspect that makes things challenging is the concept of a multi-
cloud environment. Many a time, enterprises (businesses) are required to
deploy their applications across multiple clouds that may include hybrid
cloud environments (here on-premises deployments aka a data center
leveraged alongside cloud instances).
For enterprises (businesses) that are working towards achieving digital
acceleration, this is great, however, it has the potential challenge for
cybersecurity and cloud operations teams who are required to manage
security across all cloud instances.
Solution
CSMA assists in simplification by enabling and leveraging an integrated
(single point) view across all cloud instances thereby eliminating gaps
related to visibility.
Additionally, CSMA assists in the implementation of predefined policies
related to best practices. Policies diminish the chances of human error as
they are defined consistently and monitored continuously through
automated processes. CSMA also entails transparent, holistic logging and
monitoring and is scalable based on demand.

OT security environment
Problem statement
At the core of OT is a collection of industrial control systems (ICS) which
can be a combination of sensors/monitors/other technologies used in
industrial settings. These are markedly different compared to a normal
back-office application and exhibit specific characteristics. Especially OT
generates large amounts of data continuously.
Perceived risk
Components in an ICS have implicit trust wherein if a component is
accessible, it is assumed that it is a part of the same ICS. An ICS is
explicitly required to be isolated from other systems. Now, if an ICS device
is accessible over the Internet, it no longer has the security benefits of
isolation.
Similar to the security challenges that are unique to remote work and cloud
environments, OT environments also have a distinct set of challenges that
need to be addressed.
Solution
CSMA assists with OT security best practices including asset identification
and classification. Enterprises (businesses) should prioritize the value of
different types of OT assets (some are likely to be more valuable as
compared to others). The top priority is to know what is the most important
and secure them. Equally vital is the ability to analyze traffic together with
threats and vulnerabilities. We have seen earlier in use cases sections that
even devices that are on-premises in remote work locations or the cloud
have similar requirements. CSMA helps in the reduction of the overall risk
to an enterprise (business) infrastructure and services significantly by
reducing the number of tools required to achieve security optimally.

CSMA in the healthcare industry


The evolution of digitalization in the healthcare industry is all about
advanced patient care amidst complications related to cyber threats.
Healthcare enterprises (businesses):

Function in an environment that is highly complex


Rely on a wide spectrum of third-party service providers
Depend on legacy infrastructure (outdated) and unpatched systems
(that present cyber challenges)
Need to look beyond zero trust and employ CSMA to manage
cybersecurity issues

Usage and integration of connected devices in healthcare offer immense


benefits related to patient care through improved diagnostics and
streamlined operations is no less than transformative. The connected
devices include smart medical devices, wearable health trackers, remote
patient monitoring tools, etc, and deliver data and analytics on a real-time
basis. All this enables healthcare providers to make informed decisions,
thereby enhancing patient care.
The downsides of the swift adoption of such technologies entail exposure of
the industry to cybersecurity risks. It is also important to note is the fact that
the industry has traditionally lacked investment in cybersecurity
infrastructure. Since the connected devices (as mentioned above) collect
and transmit sensitive patient data, protecting this data from breaches is
vital. Any unauthorized access can lead to identity theft/fraud or even put
patients’ lives in danger. Connected devices also, by default, lack robust
security features that make them susceptible to cyberattacks.
Equally important are the regulatory challenges wherein healthcare
enterprises (businesses) are obligated to follow stringent regulations such as
the Health Insurance Portability and Accountability Act (HIPAA) in the
US and the Network and Information Security (NIS2) Directive in
Europe. Regulations stipulate severe penalties for non-compliance.
What makes these challenges even worse is the dependency on legacy
(outdated) infrastructure. Many healthcare enterprises (businesses) have a
reliance on archaic and unsupported operating systems and software that
make them vulnerable to security vulnerabilities (since these systems are
often unpatched and not updated promptly). Also, they are often not
compatible with today’s cybersecurity solutions which leads to complicated
efforts in securing the infrastructure.
These challenges have no easy solution. Replacement of legacy
infrastructure calls for major financial investment, and many healthcare
enterprises (businesses) are restrained by budget constraints and rising
costs. Furthermore, due to the interconnections among complex systems,
standard cybersecurity solutions are not very effective.
Patient care is typically delivered from multiple locations with the
involvement of multiple actors with every actor having the requirement to
share patient data and operate in their own systems (separate and distinct).
Thus, cybersecurity solutions need to be designed accordingly to meet the
needs of the system.
This means a zero-trust architecture may not be suitable for healthcare
systems since it is not designed to manage the complex nature of healthcare
systems. For hospitals, the top priority and core competency is patient care
and not technology systems. In case they need an electronic health record
(EHR) for patients, they shall take the help of a third-party service
provider. If that EHR requires APIs from a government department or some
other data source, that source in itself will become a fourth party.
Consider these typical scenarios where a hospital is required to authenticate
an IP that has not come from a third-party EHR provider but may have
come from a fourth party. This becomes a source of breach at the hospital.
The hospital neither has any direct relationship with the fourth party nor has
any way to verify their security arrangements. They have no choice in this
situation but to have the third party do the needful.
In this scenario, zero trust does not help. Applications utilizing zero trust
will be unable to accept data from those legacy systems since these systems
are not within the zero-trust perimeter.
This means that healthcare enterprises (businesses) need to implement
CSMA that is distinctively suited to meet their demands. CSMA facilitates
the extension of security controls across assets that are widely distributed. It
is very flexible and suitable for modular approaches (viz., hybrid/multi-
cloud architectures) wherein data is stored in multiple locations (on and off-
premise).
CSMA provides control over multiple systems and solutions by bringing
them together. It builds on the already made security investments by the
enterprise (business) instead of asking for huge investments in new systems
and infrastructure.
Concepts such as zero trust, secure by design, and defense in depth need not
be discarded. Typically, the healthcare industry or other critical
infrastructure environments have legacy systems in place and these setups
rely on third and fourth-parties. Therefore, the emphasis and focus are on
interoperability and collaboration. Zero trust and defense in depth can be an
inhibitor to both.
In this increasingly interconnected world, the healthcare industry can lessen
vulnerabilities, protect patient privacy, and defend the trust of stakeholders
by implementing risk mitigation strategies and investing in innovative
security frameworks on a priority basis.
Implementation of CSMA can boost the resilience of the healthcare
industry against growing cyber threats. Enterprises (businesses) can form
robust cybersecurity defense mechanisms by decentralization of security
controls and emphasizing on identity-based access.
The need of the hour is CSMA that can adapt to the dynamic nature of
modern healthcare ecosystems.

Summarizing CSMA in healthcare


The cybersecurity challenges in the healthcare industry are unique. This is
due to a combination of factors. Firstly, the industry is a complex operating
environment with multi-stakeholders. Secondly, the industry has a
reasonable reliance on legacy infrastructure systems.
Solutions and concepts such as zero trust will not be suitable in isolation.
The need of the hour is to embrace CSMA, which can accommodate
existing cybersecurity solutions/systems and deliver robust security levels.
Also, CSMA can be a focal enabler for the widespread adoption of AI in the
healthcare industry. CSMA essentially fosters a security-centric approach
there by ensuring the integrity and confidentiality of sensitive patient data.
It enables reinforcing trust among stakeholders while facilitating the
seamless implementation of AI-driven solutions.

CSMA
The adoption of CSMA has gained positive traction, especially with the
increase in remote work arrangements and the widespread use of cloud
solutions. Enterprises (businesses) have been prompted to reassess data
access and control policies due to the evolving landscape, which has led to
the implementation of new technologies.
The increase in cyberattacks in recent years has further emphasized the
need for comprehensive cybersecurity measures, leading to the growth of
the cybersecurity mesh market.

Global market overview and growth factors


We have, over the course of all previous chapters, realized that CSMA
enables modern-day cyber defense strategy that is different (rather
transformational) from the traditional security practices, wherein there is
often reliance on a single perimeter to protect an entire IT environment.
CSMA protects each device independently, thereby aligning well with the
distributed nature of remote work and cloud environments. In short, CSMA
brings about a holistic approach to cybersecurity.
We have also noticed that the increase in remote work and extensive use of
cloud computing (that distributes devices across varied locations) has led to
a corresponding increase of momentum in the adoption of CSMA.
What has also prompted the enterprise (businesses) to reassess their data
access and control policies is the ever-evolving threat landscape. Further,
the emphasis on comprehensive and robust cybersecurity measures has been
increasing with the increase in the frequency of cyberattacks. This is the
main driver of the growth of CSMA (characterized by being more advanced
and adaptive in nature).
Demand for CSMA has been growing substantially across industry verticals
viz., banking and financial, technology, IT-enabled services, healthcare,
energy, and utilities, amongst others. This growth is testimony to the
industry's acknowledgment of the effectiveness and importance of CSMA
in addressing modern-day cybersecurity challenges.

Market dynamics
Let us now seek some of the key drivers of the adoption of CSMA.

Driver: Management of evolving threat landscape


The growth in sophistication of cyberattacks, combined with the movement
of workloads to hybrid cloud environments, is leading to an increase in
demand for CSMA. Enterprises (businesses) are increasingly recognizing
the need for a scalable and composable (modular) design to integrate
security products/tools/solutions into a collaborative ecosystem. This is
precisely what CSMA brings to the fore.
The cybersecurity landscape has been reshaped by the COVID-19 pandemic
by pushing cyber assets outside traditional logical and physical security
boundaries. As a response, enterprises (businesses) have turned to CSMA
as a practical and effective solution.
CSMA enables enterprises (businesses) by establishing security parameters
around devices and identities and providing a strong base for scalable and
adaptable cybersecurity controls. CSMA presents a strategic and future-
looking solution for enterprises (businesses) to maneuver the ever-evolving
and complex cybersecurity landscape in the modern digital world.
Overall, CSMA enables disparate technologies to communicate through
various key components, such as security intelligence, centralized policy
management, and identity fabric. These advantages over traditional siloed
cybersecurity strategies are anticipated to drive the market demand for
CSMA in the coming years.

Threat: Shortage of skilled expertise


A major challenge for enterprises (businesses) in the implementation and
adoption of CSMA is the shortage of skills. Demand for skilled (with
specialized knowledge and skills) cybersecurity workers has been on a
constant rise due to the increase in the complexity of the cybersecurity
threat landscape (which leads to growth in the number of entry points for
cyber threats).
Unfortunately, the industry is grappling with a paucity of trained
cybersecurity professionals capable of understanding and responding to
advanced cyberattacks. This shortage of trained security professionals
exposes enterprises (businesses) to heightened risks of facing cyberattacks.
Moreover, hiring and retention of a skilled cybersecurity professional is a
bigger challenge.
The increase in sophistication of cyber threats (rise of new zero-day threats)
undermines the importance of deployment of CSMA to detect and
remediate attacks. However, inadequate end-user awareness regarding
advanced cyber threats and a lack of spending on cybersecurity awareness
training contribute to the industry's skill gap.
Enterprises (businesses) are required to build specialized teams to manage
the complex threat landscape, and address the skills gap in the cybersecurity
domain is a critical imperative.
After all, the adoption and implementation of CSMA requires a specialized
and trained workforce that is skilled enough to effectively protect
enterprises (businesses) against evolving cyber threats.

Opportunity: Growing adoption of multi-cloud


While leveraging services from multiple cloud providers, enterprises
(businesses) have increasingly been adopting multi-cloud deployments.
However, this setup presents challenges related to fragmentation due to the
need to manage distributed IT assets across multiple clouds.
Management of a unified interface in a single-cloud environment is simpler
than in a multi-cloud environment, which requires management of multiple
interfaces. Understanding each component across various clouds in a multi-
cloud environment adds complexity. Moreover, establishing a consistent
security posture in a multi-cloud environment is difficult to achieve since
all cloud providers have distinct security methodologies and support
varying policies.
CSMA serves as the base for protecting all digital assets of enterprises
(businesses) by securely connecting users and compute environments over
hybrid/multi-cloud environments and across varied application
generations/channels.
Another major challenge today is the inadequacy of the existing identity
and security architectures to adapt to the evolving needs of enterprises
(businesses). CSMA manages this challenge by provisioning a
comprehensive and integrated security framework that can protect assets in
on-premises, data center, and cloud environments.
CSMA facilitates collaboration among standalone security
tools/products/solutions by standardizing the way these components
communicate with each other which leads to the strengthening of the
overall security posture.

Conclusion
Enterprises (businesses) need to adopt CSMA and build a solid baseline for
safeguarding their digital assets. This becomes increasingly important in an
ever-changing cybersecurity threat landscape. We have seen how the
implementation of CSMA presents its challenges. These can be managed
via thorough assessments, vendor collaboration, and comprehensive user
training (education).
In summary, CSMA unveils the future of cybersecurity through a holistic
and integrated approach, and enterprises (businesses) can strengthen their
cybersecurity posture by adopting CSMA.

Points to remember
As we take on board a future dotted with pervasive connectivity and
sophisticated cyber-attacks/threats, it is understood that approaches to
cybersecurity that are traditional and siloed will not serve well. The shift
towards CSMA is not just preferred, it is vital.
Toward the end of this chapter, let us summarize the practicalities of how
enterprises (businesses) can adopt this innovative approach to
cybersecurity:
Figure 9.1: Steps for adopting an innovative approach to cybersecurity

Recognize the need for change:


First vital step towards implementation
Acknowledge the constraints of current cybersecurity practices
Acknowledge the necessity for a more unified, scalable, and
flexible approach
Work towards getting a holistic evaluation done for the existing
security infrastructure and potential vulnerabilities
Partner with the right provider: Choose a befitting partner (with
recognized expertise and solid track record) such as a managed
security service provider (MSSP) that can provide built-on CSMA
Customize the solution:
Adopt and implement CSMA based on an enterprise’s
(businesses’) specific needs
The following aspects will have a bearing on the final solution -
the size of the enterprise (business), the nature of its operations,
its regulatory environment, and the specific threats it faces
Prioritize user training and education:
Comprehensive and robust cybersecurity solutions can be
weakened by human error
The need to understand the importance of cybersecurity is crucial
and needs to be emphasized through periodic user training and
education
Regularly evaluate and adapt:
The CSMA journey is not a one-time effort but an ongoing
phenomenon.
It is vital to conduct evaluations regularly. This is to enable the
assessment of the effectiveness of the existing security measures
and to make adjustments if required.
The CSMA journey in an enterprise (business) needs to adjust to
the emergence and evolution of new threats.
Future proof the journey:
This is non-negotiable for any enterprise (business) – it is vital to
choose a solution that evolves with the times (for example., swift
technological advancement in AI, ML, etc. should be widely
embraced
It is equally vital for the potential service provider to be
committed to staying at the forefront of cybersecurity
developments

The journey towards robust and unified cybersecurity may come across as
complex, but the rewards are commensurate with the effort. By adopting
CSMA, enterprises (businesses) can safeguard the present and make the
way for a safe and secure digital future.
As we conclude this book, remember that cybersecurity is a continuing and
endearing journey. A journey that necessitates constant vigilance,
continuous learning, and an inclination to adapt and evolve.

Key terms
Key performance indicators: A measure (quantifiable) of
performance over time for a specific objective.
Operational technology: The hardware and software that is used in
industrial settings to monitor and control devices, processes, and
infrastructure.
Virtual private network: A mechanism for creating a secure
connection between a computing device and a network, or between
two networks, using the public Internet.
Virtual machines: Is a physical computer’s digital version. The digital
version uses software (instead of a physical computer) to run programs
and deploy applications.
Microservices: An architectural approach to software development
wherein software comprises small independent services that
communicate over APIs.
The Open Worldwide Application Security Project: An online
community that provides free resources viz., articles, methodologies,
documentation, tools, and technologies in the fields of IoT, system
software, and web application security.
Industrial control systems: An electronic control system and
associated instrumentation used for industrial process control.
The Health Insurance Portability and Accountability Act: A
federal law (national standard) aimed at protecting sensitive patient
health information from being disclosed without the patient's consent
or knowledge.
NIS2: EU-wide legislation on cybersecurity that provides legal
measures to enhance the overall level of cybersecurity in the EU
Electronic health record: Collection of patient health information in a
digital (electronic format)

Multiple choice questions

1. What is the main focus of CSMA?


1. Centralized security model
2. Hybrid security model
3. Decentralized security model
4. Perimeter-based security model

2. Which aspect of CSMA places emphasis on the authentication and


authorization of individual users and devices?
1. Zero trust architecture
2. Adaptive security
3. Identity-centric security
4. Distributed security services

3. What security model does CSMA align with?


1. Trust-first model
2. Decentralized model
3. Centralized model
4. Zero trust architecture

4. What is the reason for CSMA prioritizing adaptive security


measures?
1. Reduce the impact of potential breaches
2. Lessen the need for continuous monitoring
3. Centralize security controls
4. Lessen the use of interconnect security fabrics

5. Owing to what characteristics does CSMA facilitate


comprehensive protection in the digital environment?
1. By isolating security elements
2. By centralizing security measures
3. Through interconnected security fabrics
4. Through security controls

6. CSMA's scalability leads to what advantage?


1. Reduced security measures
2. Organizational growth due to seamless adaptability
3. Centralized security controls
4. Limited protection of endpoints

7. Through which of the following attributes can CSMA lead to an


enhanced security posture?
1. By relying on a centralized ecosystem
2. Through a distributed approach thereby reducing single points of
failure
3. By ignoring zero trust architecture
4. By lessening the need for adaptability

8. How does CSMA manage the ever-changing cyber threat


landscape?
1. Through the adoption of static security measures
2. Through the adoption of dynamic and adaptive features
3. By way of relying on traditional network perimeters
4. Due to inadequate continuous monitoring mechanisms

9. Which of the following is the biggest reason for the success of


CSMA implementation?
1. Overlooking industry challenges
2. Getting complacent in risk assessment
3. Carrying out a holistic risk assessment
4. Lack of collaboration with vendors
10. Which vital ingredient can assist in overcoming challenges in
CSMA implementation?
1. Lack of user awareness and training
2. Competition amongst vendors
3. Governance (collaborative) involving key stakeholders
4. Reliance on a single security measure

Answer key

1. Decentralized security model: CSMA moves away from


outdated security models that are centralized, and distributes
security controls across the digital ecosystem, thereby
making a provision for a decentralized and adaptive defense
ecosystem.
2. Identity-centric security: CSMA does not only rely on
traditional network perimeters but also emphasizes
authentication and authorization of individual users and
devices thereby enhancing visibility and control.
3. Zero trust architecture: CSMA has a zero-trust security
model at its core. This requires continuous verification (for
every user/device/application).
4. Reduce the impact of potential breaches: CSMA uses
adaptive security that facilitates the adjustment of security
controls on a real-time basis. This leads to a reduction in the
impact of potential security breaches.
5. Through interconnected security fabrics: CSMA
propagates that interconnected security fabrics be created.
This facilitates collaboration and provides holistic protection
across the entire digital ecosystem.
6. Organizational growth due to seamless adaptability: A
key component of CSMA is scalability. This facilitates
seamless adaptability to the expansion of digital ecosystems.
This further ensures that security measures do not hinder
organizational growth.
7. Through a distributed approach thereby reducing single
points of failure:
CSMA facilitates the dispersion of security controls. This
leads to the minimization of the risk of a single point of
failure and enhances overall security posture.
8. Through the adoption of dynamic and adaptive features:
The basic nature of CSMA is its ability to dynamically adapt
which enables it to respond in a real-time basis to emerging
threats in this ever-evolving cyber threat landscape.
9. Carrying out a holistic risk assessment: A comprehensive
risk assessment is vital for the identification of
vulnerabilities and prioritizing security measures during the
adoption of CSMA.
10. Governance (collaborative) involving key stakeholders:
The involvement of key stakeholders (IT, security, and
business units) in governance (collaborative) enables the
comprehensive implementation of CSMA.

Join our book’s Discord space


Join the book's Discord Workspace for Latest updates, Offers, Tech
happenings around the world, New Release and Sessions with the Authors:
https://discord.bpbonline.com
Index
A
Advanced Research Projects Agency Network (ARPANET) 4
Anonymous 7
anti-virus 4
application programming interface (API) 92
artificial intelligence (AI) 51

B
Business Email Compromise (BEC) 36

C
California Consumer Privacy Act (CCPA) 55
chief information security officers (CISOs) 44
Cloud Access Security Broker (CASB) 94
cloud attacks 40
Cloud Security Posture Management (CSPM) dashboard 96
Cloud Service Provider (CSP) 40, 51, 78
commercial antivirus 5
computer crime 5
computer virus 5
Creeper 4
cryptojacking attacks 34
CSMA adoption 170
global market overview 171
growth factors 171
market dynamics 172-174
CSMA adoption, key aspects 113
automation 114, 115
comprehensiveness 114
integration 113, 114
CSMA components 92, 93
automated operations 95
centralized operations 95
comprehensive security intelligence 94, 95
consolidated dashboards 96
distributed identity fabric 94
important component 96
integrated data 93, 94
CSMA in healthcare industry 167-170
CSMA solutions 134
AI graph 134
data normalization and correlation 134
generative AI for context enhancement 135
natural language processing (NLP) 135
cyber resilience 64
building blocks 65, 66
importance 64, 65
cybersecurity
building blocks 18
impacting, by major shifts 15-17
traditional measures 19, 20
cybersecurity attacks 31
cloud attacks 40, 42
cryptojacking attacks 34, 35
Denial-of-Service attack 42, 43
injection attacks 43
insider attacks 41
Internet of Things (IoT) attacks 38
malware attacks 31, 32
man-in-the-middle (MitM) attacks 41
Phishing attack 36
ransomware attacks 33, 34
social engineering attacks 36
state-sponsored attacks 40
supply chain attacks 35, 36
cybersecurity ecosystem
current situation 70, 71
cybersecurity events
noteworthy events of 2013 11
noteworthy events of 2014 11, 12
noteworthy events of 2015 and 2016 12
noteworthy events of 2017 12, 13
noteworthy events of 2018 13
noteworthy events of 2019 14
noteworthy events of 2020 14
noteworthy events of 2021-2023 and beyond 14
cybersecurity evolution 2
ARPNET 4
commercial antivirus 5
connected world 6
Creeper 4
first DOS attack 5
Reaper 4
time before networking 2
timeline view 10
time of early hacking 3, 4
time of email 6
time of phone phreaks 3
cyber security future trends 58
5G network security 60
AI-assisted security testing 61
AI-driven cybercrime 62
common themes 62-64
cryptography resistant to quantum computing 60
cyber insurance 61
ethical hacking and bug bounty programs 62
human-centric security 60, 61
smart cities and critical infrastructure security 62
stronger authentication measures 60
zero trust 59, 60
cybersecurity key issues 44
artificial intelligence issues 45, 46
budgets issue 44, 45
skills gap issue 45
cybersecurity landscape today 110, 111
adoption of cloud 112
attack surface expansion 111, 112
Cybersecurity Mesh Architecture (CSMA) 71
adoption considerations 119-122
befitting vendor(s) 118
benefits 78-85
benefits of leveraging 127-131
challenges 73-78, 158, 159
characteristics 131, 132
comparing, with other architectures 152, 153
compatible tools, evaluating 117, 118
consolidation 118
current tools and identify gaps, assessing 116
implementing 150-152
key components, building 116, 117
key performance indicators 153, 154
layers, illustrating 71, 72
necessity 127
need for 72
performance 108-110
pitfalls, of not leveraging 135-140
products/solutions 98-102
real-life scenario 149, 150
review and revise 119
revisit 90
salient points and goals 147, 148
scalability relative, to other architectures 92
target use cases 133
ten commandments 154-158
unified architecture 96, 97
versus, defense-in-depth approach 144, 145
working 92
working relationships, with zero trust 91
working with 115
cybersecurity threat canvas expansion 6, 7
Anonymous 7
artificial intelligence, using 9
GDPR 8
Operation Aurora 8
remote and hybrid work 9
SolarWinds attack 9
Stuxnet 8
X hack 8
cybersecurity threats 29, 30
evolution 17
Life Stages (LS) 17, 18
sources 30
cybersecurity trends 49, 50
AI power 53
board room agenda 58
continued prominence of regulations 57, 58
cyber resilience 58
cyber warfare, intensifying by rogue nations 53
data privacy 55
Deepfakes threats 56, 57
emphasis on incident response and recovery 58
enhanced automation 54
imminent cloud security threats 51, 52
IoT threats 53
IT skills gap widening 56
multi-factor authentication (MFA), improving 54
rise in insider threats 56
risks of distributive workforce 55, 56
smarter social engineering attacks 51
supply chain attacks 57
tenacious data breaches 50
threat emanating due to mobile devices 52
upsurging ransomware 51

D
Denial-of-Service attack 42
denial of service (DoS) 30
Disaster Recovery as a Service (DRaaS) 34
Distributed Denial of Service (DDoS) attacks 32
distributed systems 26, 27
security challenges 27-29
domain name server (DNS) 42

E
early hacking 3
Early Internet 4
electronic health record (EHR) 169
ethical hacking 4
European Union (EU) 8
F
first DOS attack 5

G
General Data Protection Regulation (GDPR) 8, 55
Generative AI (GenAI) 38
phishing attacks 38

H
hacking 3
Health Insurance Portability and Accountability Act (HIPAA) 168
HTTP flood DDoS 42
HTTPS spoofing 42
HyperText Transfer Protocol Secure (HTTPS) 6

I
ICMP flood 43
Identity and Access Management (IAM) 18
Infrastructure as a Service (IaaS) 90
injection attacks 30, 43
vectors 43, 44
insider attacks 41
Insider Threat 8
Internet Chat Relay (IRC) channel 7
Internet of Things (IoT) attacks 38
privacy concerns of connected cars 39
smart medical devices and electronic medical records vulnerabilities 39
Internet Protocol spoofing 42
Intrusion Prevention Systems (IPS) 17

K
key performance indicators, CSMA 153, 154

L
lightweight directory access protocol (LDAP) 44

M
machine learning (ML) 51
machine learning (ML) algorithms 9
malicious software 31
malware attacks
adware 32
cryptojacking 32
fileless malware 32
ransomware 32
spyware 32
trojans 32
viruses 32
worms 32
Managed Service Providers (MSPs) 35
man-in-the-middle (MitM) attacks 30, 41
examples 41, 42
Mean Time To Detect (MTTD) 28, 75
Mean Time To Remediate (MTTR) 28, 75
Micro, Small, and Medium Enterprises (MSMEs) 40
modular 92
multi-factor authentication (MFA) 9, 16

N
natural processing language (NLP) 53
Network and Information Security (NIS2) Directive 168
network time protocol (NTP) 43
next generation set 70

O
operating system (OS) security 7
Operation Aurora 8
OT security environment 166, 167

P
payload malware 18
Personally Identifiable Information (PII) 15, 55
Phishing attack 36, 37
phone phreaks 3
polymorphic viruses 6

R
Ransomware-as-a-Service (RaaS) 34
ransomware attacks 33, 34
Reaper 4
return on investment (ROI) 44

S
Secure Sockets Layer (SSL) protocol 6
Security Information and Event Management (SIEM) 28
Security Operations (SecOps) 26
social engineering attacks 36
Software as a Service (SaaS) 90
SolarWinds attack 9
SolarWinds Supply Chain Attack 9
spear-phishing 36
state-sponsored attacks 40
Structured Query Language (SQL) query 43
Stuxnet 8
supply chain attacks 35
SYN flood DDoS 42

T
target use cases, CSMA
compliance management 134
security operations 133
vulnerability management 133
The Orange Book 5
Trojan horse 5

U
unified architecture, of CSMA 96, 97
accelerated through automation 98
extensive 97
integrated 97, 98
unique identifier (UID) 38
user datagram protocol flood DDoS 43
user datagram protocol (UDP) 43

V
virtual machines (VMs) 165
virtual private networks (VPN) 162
vishing 37

W
WannaCry ransomware (2017 event) 16
whaling 37
work-from-home environment 162
endpoint detection and responses 163, 164
home network security 164-166
zero-trust network access 162, 163

X
X hack 8

Z
zero day attack 7
Zero Trust Architecture (ZTA) 59

You might also like