Information Systems for Managers (ISM)
Cloud and Big Data
Session 5
Cloud Computing
Image Source: D. Fletcher for cloudTweaks.com
2
https://www.youtube.com/watch?v=AnxrJiS5uKU 3
What is
Cloud
Computing?
4
Some Statistics
Cloud market projected to reach $864 bn. in 2025
5
Cloud Computing
• Cloud computing is a model for
enabling convenient, on-demand
network access to a shared pool of
configurable computing resources (e.g., CLOUD:
C – Common Infrastructure
networks, servers, storage, L – Location Independence
O – Online Accessibility
applications, and services) that can be U – Utility Pricing
D – on Demand resources
rapidly provisioned and released with
minimal management effort or service
provider interaction. [NIST]
• Cloud Computing enables computing
facilities and infrastructure to be made
available over internet. 6
But Why?
Capacity expansion
through IT asset
purchases results in large
capital investments
7
Key technologies behind cloud computing
Virtualization Multitenancy Web Services
• Virtualization is technology Web Service is a strategy to
• Virtualization + Resource-
that allows you to create sharing = Multitenancy make the services of one
multiple simulated application available to
environments or dedicated • Multitenancy allows other applications using
resources from a single, multiple users (tenants) to messaging protocols (read
physical hardware system. work in the same “no user interface).
• Software called a software environment at
hypervisor connects the same time at their Microservice is a completely
directly to that hardware own user interfaces autonomous application that
(skins). performs a specific service in
and allows you to split one
system into separate, • SaaS (Software-as-a- a larger application
distinct, and secure Service) uses a highly architecture.
environments known as multi-tenant architecture.
virtual machines (VMs). E.g. Google Apps
Capabilities
Scalability Elasticity Measured Use Usage-Based Pricing
• Ability address needs of • Ability to expand and • Includes metrics and all • Can be based on usage and
increase demand contract depending on the services are measured needs of client
demand throughout
Managed Services Service Levels Ubiquitous Access Heterogeneity
• Building, maintain and • Duration of time an infra • Anytime-anywhere access • Can be managed easily
upgrading the system is done service to be available w/o through virtualization
by provider interruption or the speed or
response level of service
Reduced Setup Time Resource Pooling
• No need to purchase • Infra is created with hundred
hardware or do civil works to or thousands of CPUs and
setup memory device and only a
slice of it is allotted to client
9
Application Infrastructure
10
Service Models
Operating Environment with
Applications
End Users
e.g., CRM, ERP
SaaS
Development platforms
e.g., Google App Engine Application Developers
PaaS
Storage and Compute
e.g., Server, Database, Network Infra. Admins
IaaS
11
Cloud Computing Service Models
Infrastructure Platform Software
(as a Service) (as a Service) (as a Service)
You Manage
You Manage
Applications Applications Applications
Security and Security and Security and
Integration Integration Integration
Managed by vendor
Databases Databases Databases
Managed by vendor
Vertualization Vertualization Vertualization
Managed by vendor
Server OS Server OS Server OS
Storage Storage Storage
Networking Networking Networking
Server HW Server HW Server HW
SaaS v/s PaaS v/s IaaS
SaaS PaaS IaaS
• On demand service • This service is made up of a OS, a • Offers computing architecture
• Platform independent web server, and a database and infrastructure in virtual
• No need to install the s/w • Users build, compile, and run environment
on your PC programs without worrying • Multiple users can access four
• Used by multiple users about underlying infrastructure resources, managed by vendors:
• Managed by vendor • You manage data and application Data Storage, virtualization,
resources Servers and Networking
User: End users Users: Developers Users: System Administrators
Examples: Gmail, MS office Examples: Google App Engine, Examples: Amazon S3 storage
365, Google Drive, CRM, Microsoft’s Azure Services Platform, service, Rackspace cloud servers; EC2
ZOHO, Force.com LAMP platform
Adv: Adv: Adv:
• Universally accessible • Cost effective rapid development • Enhanced Scalability
from any platform • Easy development of web app • IaaS is flexible
• Excellent for collaboration
Disadv: Disadv: Disadvantages
• Performance based on • Migration issues: the risk of • Security issues
internet speed vendor lock-in • Network and service delays
• Browser issues
Cloud Computing Deployment Models
Private Cloud Public Cloud Hybrid Cloud Community Cloud
• A truly shared • Private Cloud with • Shared Computing
• Single Client
facility public resources infra
• Leverages Existing
• Shifts CapEx to • Address problem • Maybe setup by a
CapEx
OpEx of Cloud ‘bursts’ group of companies
• Can Reduce OpEx
• Offers a Pay as you • Infra is provided in sharing similar
• Though security is real time from
go model specifications
key feature lacks public cloud
core advantages of • Supports Multiple • E.g.: Bank, Hospitals
CC Tenants etc
Public vs. Private Cloud Deployment
Public Private
Economies of Scale Control
Cost Security
Flexibility Regulation
Op-ex Cap-ex
Players in the Cloud Industry
Provider Enablers Consumers Regulators
• Own and • Companies • Corporate or • Government or
manage their who facilitate individuals international
own systems – the cloud bodies
deliver ‘cloud’ client-provider
needs through partnership
cloud services through
migration,
monitoring,
maintenance,
etc
16
Cloud Providers
Gartner Magic Quadrant CSP 2009 Gartner Magic Quadrant CSP 202217
Amazon Web Services (AWS)
Market leader in cloud services (IaaS,
PaaS)
Offers more than 250 cloud-based services
Provide cloud solutions to millions of users
30 regions
96 availability zones
245 countries and territories served
Computing
EC2 – Micro, small, medium, large
On-demand ($0.05/hour), reserved (72%
discount), spot (90% discount)
18
Salesforce (SaaS)
It provides CRM software and applications focused on sales, customer
service, marketing automation, analytics, and application development.
#1 CRM: More than 1,50,000 clients use Salesforce’s CRM application
Sales Cloud, Service Cloud, Platform, Marketing Cloud, Commerce
Cloud, Community Cloud
Unique aspects of Salesforce.com services include ease of use,
customizability, continuous improvement & innovation
19
CRM
20
BUSINESS VALUE OF CLOUD
COMPUTING
21
Business value of cloud computing
Financial Drivers Operational Driver Strategic Drivers
• Reduced total • Agility • Develop
cost of • Scalability capabilities
ownership • Centralized • Rapid product
• Opportunity cost management development
• Productivity • Increased access • Customer
intimacy
22
When not to move to cloud?
• Tight coupling between hardware and
software
• Applications that do not scale
• The cost not justified
• IT staff may not have the relevant skills
• Security is a key concern
23
Applications that have volatile demand
Low business risk and commodity
application
When to
move to Applications that require collaboration
cloud?
Big data storage and analysis
Capital available for infrastructure is low
19-01-2024 ITS S12 | Vimal Kumar M | IIM Calcutta 24
Challenges
Privileged User Access
Regulatory Compliance
Data Location
Recovery
Investigative Support
Long-term Viability
Benefits
Significantly lower the cost of entry for newer firms that lack resources
Immediate access to hardware resources without capital investment leads to faster
time to market
Lowers IT barriers to innovation, helps organizations in digital transformation
Capabilities of cloud enables scale their business capabilities based on demand
Facilitates the development of new class of applications – mobile interactive
applications and batch processing
26
BIG DATA AND ANALYTICS
27
Big Data Definition
• No single standard definition…
“Big Data” is data whose scale, diversity, and
complexity require new technologies,
algorithms, and analytics to manage it and
extract value and hidden knowledge from it…
28
Characteristics of Big Data:
1-Volume
• Data Volume
– 44x increase from 2009 2020
– From 0.8 zettabytes to 35zb
• Data volume is increasing
exponentially
500 million tweets per day
350,000 tweets per minute
6,000 tweets per second
Exponential increase in
collected/generated data 29
30
Characteristics of Big Data:
2-Variety
• Various formats, types, and
structures
• Text, numerical, images,
audio, video, sequences,
time series, social media
data, multi-dim arrays, etc…
• Static data vs. streaming data
• A single application can be
generating/collecting many
types of data
To extract knowledge all these types of data need to linked together
31
Characteristics of Big Data:
3-Velocity
• Data is generated fast and need to be
processed fast
• Online Data Analytics
• Late decisions missing opportunities
• Examples
– E-Promotions: Based on your current location, your purchase history,
what you like send promotions right now for store next to you
– Healthcare monitoring: sensors monitoring your activities and body
any abnormal measurements require immediate reaction
– LinkedIn: People You May Know, Groups You May Like, Jobs You May Be
Interested In
32
Big Data: 3V’s
33
Data Vs Big Data
Structure
d
Some Make it 4V’s
35
Or (5V)
Customer BD: 1 TB = 220 DVDs
sentiments are 16m FB photos
not verifiable
Text, photo, audio,
video, clickstream,
sensor data
LinkedIn, Amadeus (travel), GE
gas turbine
Batch Real-time Revenue , Operational costs
20b IoT devices by 2020 Serve customers better
Additional Dimensions…
• Variability: Variation in the data flow rate
• Complexity: Number of data sources
• Big data are collected from numerous data sources
• Complexity makes it difficult to collect, clean, store, and
process heterogeneous data
• Decay: Declining value of data over time
• In a time of high velocity, the timely processing and action is
the key
• Example: Data generated by wearable medical devices
Sources of Big Data
38
Numerous possibilities
Style of data Source Industry affected Function affected
Large volume Online Financial services Marketing
Unstructured Video Health care Supply chain
Continuous flow Sensor Manufacturing Human resources
Multiple formats Genomic Travel/transport Finance
Activity: Identify one application of big data by combining the dimensions
39
Applications Starbucks introducing new Drilling for Oil Chevron
coffee products
Monitoring trucks at US Target targeting customers
express
https://www.youtube.com/watch?v=XH1wQEgROg4 40
The Model Has Changed…
• The Model of Generating/Consuming Data has Changed
Old Model: Few companies are generating data, all others are consuming data
New Model: all of us are generating data, and all of us are consuming data
41
Difference between traditional data
and big data?
Continuous approach to sampling, analyzing
and action taking is necessary
Data is constant state of change; discovery and
agility is critical, not stability
42
Difference between traditional data
and big data?
Big data Traditional data
Type of data Unstructured format Formatted in rows
and columns
Volume of data 100 TB to PB Tens of TB or less
Flow of data Continuous flow of Static flow of data
data
Analysis methods Machine learning Hypothesis-based
Primary purpose Data-based products Internal decision
support and analysis
43
Analyzing data
Term Time frame Meaning
DSS 1970-1985 Use of data analysis to support decision making
EIS 1980-1990 Focus on data analysis for decisions by senior executives
OLAP 1990-2000 Software for analyzing multi-dimensional data tables
BI 1989-2005 Tools to support data driven decisions with emphasis on
reporting
Analytics 2005-2010 Focus on statistical and mathematical analysis for
decisions
Big data 2010- Focus on very large unstructured, fast-moving data
44
An Integrated View of Big Data
Processes for extracting insights from
Big Data
Stage 1: Acquisition Extraction, Integration,
Data and Cleaning and Aggregation and
Management Recording Annotation Representation
Stage 2: Modeling
Interpretation
Analytics and Analysis
Big Data Technology
47
Map Reduce Data Flow
48
Big Data Technology
Technology Definition
Hadoop Open-source software utilities for
processing big data across multiple
servers
MapReduce The architectural framework on which
Hadoop is based
Scripting languages Programming languages that work well
with big data (e.g., Python, Pig, Hive)
Machine learning Algorithm for rapidly finding the best
model fit of data set
Visual analytics Data Visualization
Natural language processing (NLP) Software for analyzing the text-
frequencies, meanings, etc.
49
Stage 2: Analytics
1
• Text Analytics
2
• Audio Analytics
3
• Video Analytics
4
• Social Media Analytics
5
• Predictive Analytics
Big Data Analytics Cycle
Data
Action Insights
Decision
Thank You
52
