CHAPTER 2: FOUNDATION

AND CLASSIFICATION
THS. NGUYỄN ĐÌNH THỌ
[email protected]
01678868285
1
CONTENTS
Data Center

Virtualization technology

Model Services
Market-oriented
architecture
Simulation Toolkits 2
WHAT IS DATA CENTER ?

3
 WHAT IS DATA CENTER?
• Define of data center (DC) • Trung tâm dữ liệu là một cơ sở được sử
Data centre is a facility used to house computer dụng để chứa các hệ thống máy tính
systems and associated components, such và các thành phần liên quan, chẳng
as telecommunications and storage systems. It hạn như hệ thống viễn thông và lưu
generally includes redundant or backup power trữ. Nó thường bao gồm nguồn cung
supplies, redundant data communications
cấp dự phòng, kết nối truyền dữ liệu
connections, environmental controls (e.g. air
dự phòng, điều khiển môi trường (ví
conditioning, fire suppression) and various
security devices. A large data center is an
dụ: điều hòa, triệt tiêu lửa) và các
industrial-scale operation using as much thiết bị bảo mật khác nhau. Một trung
electricity as a small town. tâm dữ liệu lớn là một hoạt động quy
mô công nghiệp sử dụng nhiều điện
https://en.wikipedia.org/wiki/Data_center như một thị trấn nhỏ.
4
DATA CENTER ARCHITECTURE

5
VALUE OF YOUR DATA CENTER

• Data Centers are specialized environments that safeguard your

company's most valuable equipment and intellectual property.
• Data Centers house the devices that do the following:
• Process your business transactions
• Host your website
• Process and store your intellectual property
• Maintain your financial records
• Route your e-mails
6
VALUE OF YOUR DATA CENTER-BRAIN
• Your business' ability to:
• Perceive the world (data connectivity)
• Communicate (e-mail)
• Remember information (data storage)
• Have new ideas (research and development)

• How to:
• Secure the brain
• Help it function efficiently
• Develop its full potential for your business
7
DATA CENTER COMPONENTS

Basic Data Center facility systems:

• Physical space
• Raised flooring
• In-room electrical
• Standby power
• Data cabling
• Cooling
• Fire suppression
8
DATA CENTER COMPONENTS

• Physical Space
• Physical space refers to the footprint that Data Center-related items
occupy. This generally applies to the overall area of the Data Center
and its associated spaces, such as electrical rooms or storage areas.
On a smaller scale this might refer to key dimensions within the Data
Center, such as the external measurements of a server cabinet or
aisle clearances.

9
DATA CENTER COMPONENTS
• Raised Flooring
• Raised flooring is an elevated grid system that is frequently installed in large Data
Centers. Cooled air, electrical whips, and data cabling are routed through the space
under the raised floor, promoting better air flow and enabling easier management of
power and cable runs. Water pipes, fire suppressant cylinders, moisture detectors, and
smoke detectors may be located here as well.
• Raised flooring can vary in height from a few inches to several feet, or a few
centimeters to several meters. In extreme cases they are as tall as the story of a
building, enabling workers to walk upright under the plenum. Regardless of their height,
the floors are typically composed of standard 2 foot (60 centimeter) square floor tiles.
The tiles can vary in weight, strength, and finish depending upon their use. Tiles
featuring either small perforations or large cut-out sections are placed in key locations
to enable pass-through of air and cabling between the areas above and below the floor.
10
DATA CENTER COMPONENTS

• In-Room Electrical
• In-room electrical refers to all power-related facilities within the Data
Center. This normally includes electrical panels, conduits, and several
types of receptacles. Power to this system usually comes from an
outside commercial power source, namely your local utility company,
and is likely conditioned at the company site. Voltage varies from one
country to another.

11
DATA CENTER COMPONENTS

• Standby Power
Standby power includes all backup power systems responsible
for support of the Data Center's electrical load in the event
that normal utility power fails for any reason. This system
traditionally includes large batteries, known as an
uninterruptible power source or uninterruptible power supply,
and one or more generators.

12
DATA CENTER COMPONENTS

• Cabling
The cabling system is all structured cabling within the Data Center.
Copper and fiber cabling are the typical media and are terminated via
several types of connectors. Common components include fiber
housings, patch panels, multimedia boxes, and data faceplates.
Cabinets, raceways, and other items used to route structured cabling
are also considered part of the cabling system. Users plug servers in to
the Data Center's structured cabling system with pre-terminated patch
cords.
13
DATA CENTER COMPONENTS

• Cooling
The cooling system refers to the chillers and air handlers used
to regulate ambient temperature and control humidity within
the Data Center. This system might incorporate the air
conditioning system used to cool regular office space within
the same building, known as house air, or might be
independent of it. Individual server cabinets can also possess
their own cooling measures, such as fans or water-cooling.
14
DATA CENTER COMPONENTS

• Fire Suppression
Fire suppression includes all devices associated with detecting or extinguishing a
fire in the Data Center. The most obvious components are water-based sprinklers,
gaseous fire suppression systems, and hand-held fire extinguishers. Others can
include devices that detect smoke or measure air quality.
• Other Infrastructure Components
There are also some infrastructure items that do not strictly fall under the prior
categories but are commonly found in server environments. These include leak
detection devices, seismic mitigation, and physical security controls such as card
readers and security cameras.
15
DATA CENTER MODELS

• Basic

16
DATA CENTER MODELS

• Multiple Conection

17
DATA CENTER MODELS

• Simple

18
DATA CENTER MODELS

• Optical fiber

19
LEVELS

20

https://www.netari.com/single-post/2014/02/04/What-to-
Look-for-in-a-Data-Center-Understanding-Tier-Levels-
DATA CENTER DESIGN CRITERIA

• How many layers of infrastructure should your Data Center possess?

• Will it be the only server environment for your company or one of several?
• Will the room house production servers and be a business-critical site or
contain a minimum of equipment for disaster recovery purposes and serve
as a failover location?
• How long is its initial construction expected to meet your company's
needs?
• What is it all going to cost?
21
DATA CENTER DESIGN CRITERIA

• Availability
• Infrastructure Tiers
• One Room or Several?
• Life Span
• Budget Decisions

22
TRADITIONAL DATA CENTER

• There are n applications stored in the network

• Each application has its own physical store
• Increased application means increased physical storage
• Complexity increases

23
MAIN FEATURES OF TRADITIONAL DATA
CENTERS

• Store data in your organization's intranet

• Requires special equipment and IT knowledge
• Hardware environment is not uniform
• The workload is complex
• The combination of different software architectures
• Use different management tools
24
DATA CENTER ARCHITECTURE IN THE
CLOUD

• There are n applications hosted in different locations that

reside on the same "cloud".
• The primary and backup databases reside on the same cloud.
• There is no loss of data when the primary database goes
down.

25
MAIN FEATURES OF CLOUD DATA
CENTERS

• Store data in Internet

• None Requires special equipment and IT knowledge
• Hardware environment is uniform
• The workload is simple
• The combination of software architectures
• Use standard management tools
26
COMPARING TRADITIONAL DATA
CENTER AND CLOUD DATA CENTER ??

27
COMPARING TRADITIONAL DATA
CENTER AND CLOUD DATA CENTER ??

28
HOW MUCH DOES A DATA CENTER COST
TO RUN?

• It depends on these things:

 How big it is.
 Where it is.
 What it’s doing.

29
THE COST OF RUNNING A TRADITIONAL
DATA CENTER
• Although each data center is a little different, the average cost per year to
operate a large data center is usually between $10 million to $25 million.
• Where’s the bulk of the money going?
- 42 percent: Hardware, software, disaster recovery arrangements,
uninterrupted power supplies, and networking.
- 58 percent: Heating, air conditioning, property and sales taxes, and labor
costs. (In fact, as much as 40 percent of annual costs are labor alone).
- Some estimates show 80 percent of spending on maintenance.
30
WHY NEED CLOUD DATA CENTER ?

• Most data centers run a lot of different applications and have

a wide variety of workloads.
• Many of the most important applications running in data
centers are actually used by only a relatively few employees.
• Some applications that run on older systems are taken off the
market (no longer sold) but are still necessary for business.

31
THE COST OF RUNNING A CLOUD DATA
CENTER

• Cloud data centers are:

Constructed for a different purpose.
Created at a different time than the traditional data center.
Built to a different scale.
Not constrained by the same limitations.
Perform different workloads than traditional data centers.
32
THE COST OF RUNNING A CLOUD DATA
CENTER

• Labor costs are 6 percent of the total costs of

operating the cloud data center.
• Power distribution and cooling are 20 percent.
• Computing costs are 48 percent.
https://www.dummies.com/programming/cloud-compu
ting/comparing-traditional-data-center-and-cloud-data
-center-operating-costs 33

/
CHALLENGES CLOUD DATA CENTER

34
CHALLENGES DC

35
VIRTUALIZATION TECHNOLOGY

• Virtualization Defined:
In computing, virtualization refers to the act of creating a
virtual version of something, including virtual computer
hardware platforms, storage devices, and computer network
resources.

https://en.wikipedia.org/wiki/Virtualization
36
VIRTUALIZATION ARCHITECTURE

37
THE BENEFITS OF VIRTUALIZATION-
DISCUSS GROUPS!

38
THE BENEFITS OF VIRTUALIZATION

• Simple management
• Fast deployment
• Quick Recovery and System Storage
• Balance and resource allocation flexibility
• Cost savings
• Enhance continuity, limit interruptions
39
VIRTUALIZATION ARCHITECTURE

• Hosted-based

40
VIRTUALIZATION ARCHITECTURE

• Bare-Mental hypervisor

41
VIRTUALIZATION ARCHITECTURE

• Hybrid

42
VIRTUALIZATION CHALLENGES

43
VIRTUALIZATION CHALLENGES

• Mass data loss: If some calamity hits the datacentres then it

might destroy the data stored in the datacentres or might
shutdown servers.
• Infected application: If a virus is infects one file then it may
corrupt whole system.
• Data integrity: The integrity of data can be affected as anyone
can access it from anywhere
http://dx.doi.org/10.4172/2165-7866.1000136
44
MODEL SERVICES

45
ADVANTAGES OF SOFTWARE AS A
SERVICE??

46
SAAS-ADVANTAGES OF SOFTWARE AS A
SERVICE

• Reduced time to benefit

Different from the traditional model, in SaaS the software
(application) is already installed and configured. The user has
the advantage of provisioning the server for an instance in
cloud and in a couple hours they can have the application
ready for use. This reduces the time spent in installation and
configuration, and can reduce the issues that can get in the
way of the software deployment.
47
SAAS-ADVANTAGES OF SOFTWARE AS A
SERVICE
• Lower cost
SaaS has a differential regarding costs since it usually resides in a shared or
multitenant environment where the hardware and software license costs are
low compared with the traditional model.
Another advantage is that the customer base can be increased since it allows
small and medium businesses (SMB) to use a software that otherwise they
would not use due to the high cost of license.
Maintenance costs are reduced as well, since the SaaS provider owns the
environment and it is split among all customers that use that solution
48
SAAS-ADVANTAGES OF SOFTWARE AS A
SERVICE

• Scalability and integration

Usually, SaaS solutions reside in cloud environments that are
scalable and have integration with other SaaS offerings.
Comparing with the traditional model, users do not have to
buy another server or software. They only need to enable a
new SaaS offering and, in terms of server capacity planning,
the SaaS provider will own that.

49
SAAS-ADVANTAGES OF SOFTWARE AS A
SERVICE

• New releases (upgrades)

SaaS providers upgrade the solution and it becomes available
for their customers. Costs and effort associated with upgrades
and new releases are lower than the traditional model that
usually forces the user to buy an upgrade package and install
it, or pay for specialized services to get the environment
upgraded.

50
SAAS-ADVANTAGES OF SOFTWARE AS A
SERVICE

• Easy to use and perform proof of concepts

SaaS offerings are easy to use since they already come with
best practices and samples inside it. Users can do proof of
concepts and test the software functionality or a new release
feature in advance. Also, they can have more than one
instance with different versions and do a smooth migration.
Even for large environments, users can use SaaS offerings to
test the software before buy it.
51
TRADITIONAL SOFWAVE VS SAAS ??

52
ADVANTAGES OF PLATFORM AS A
SERVICE??

53
PAAS-ADVANTAGES OF PLATFORM AS A
SERVICE
• REDUCING COSTS
 Server and storage overhead:
 Network bandwidth:
 Software maintenance
 Support personnel
 Careless mistakes
 Lower skill requirements:
• IMPROVING SPEED, FLEXIBILITY, AND AGILITY
54
ADVANTAGES OF PLATFORM AS A
SERVICE

• Server and storage overhead:

With a PaaS, companies don’t have excess resources in
reserve. The development tools are provided by the PaaS, and
not all of the iterations of the code need to be stored in the
data center. These savings on server and storage overhead are
realized whether developing on-premises or in the cloud.

55
ADVANTAGES OF PLATFORM AS A
SERVICE
• Network bandwidth:
The development and deployment process can put a strain on
network bandwidth within a data center. Development teams must
perform workload testing to see how the application will perform
under different circumstances. This requirement to allocate network
resources may slow down the operation of other applications or may
require the acquisition of more bandwidth capacity. PaaS enables
testing to be done in the cloud, rather than in the data center.
56
ADVANTAGES OF PLATFORM AS A
SERVICE
• Software maintenance
The cost of managing software updates and changes is often a burden
to development and operational organizations and a huge expense in
terms of time and money. Although the cost structure of a PaaS requires
a per-user, per-month charge, the cost is typically offset by reducing or
eliminating software license costs and yearly maintenance fees. The
platform vendor manages all patches and updates for the hardware and
software and also provides physical and software security for the
automation of day-to-day tasks.
57
ADVANTAGES OF PLATFORM AS A
SERVICE
• Support personnel
To keep software and systems up to date and running smoothly and to
fix problems when they occur, organizations must have IT staff at the
ready — for everything from storage and archiving to patch
management, networks, security, and the help desk. By adopting a
standardized platform across an organization, hardware and software
conflicts are greatly reduced, resulting in simplified service and support.
This level of standardization and automation allows organizations to
reduce or refocus its teams away from routine tasks
58
ADVANTAGES OF PLATFORM AS A
SERVICE

• Careless mistakes
During application development and deployment, there’s also
a great deal of pressure on teams to get work done quickly.
This pressure often results in careless mistakes. With PaaS,
such mistakes are reduced or entirely eliminated because the
platform has been fully tested and is known to work.
Developers don’t get tied down with the middleware and
tedious tasks that are prone to hasty mistakes.
59
ADVANTAGES OF PLATFORM AS A
SERVICE
• Lower skill requirements:
Development tools and middleware are complex and aren’t
standardized. Successfully deploying an application takes a high
degree of skill and experience. The learning curve on these skills is
steep, and there’s also an ongoing need to manage these components.
By providing the development tools and middleware, a PaaS lowers the
skill level required to deploy applications and removes the bottleneck
that can form while waiting for one specific person’s assistance.

60
ADVANTAGES OF PLATFORM AS A SERVICE

• IMPROVING SPEED, FLEXIBILITY, AND AGILITY

PaaS can provide greater speed, flexibility, and agility to the
development process. By providing a predictable,
heterogeneous application infrastructure, organizations don’t
get bogged down with enabling applications and can quickly
meet the needs of customers.

61
ADVANTAGES OF PLATFORM AS A SERVICE

• IMPROVING SPEED, FLEXIBILITY, AND AGILITY

PaaS helps to do the following:
Enable faster time to market by allowing development teams to focus on the
application
Enhance ability to react to changes and opportunities because the
organization doesn’t have large up-front costs associated with typical
application development and deployment
Spread capital investments further, which allows a company to be more
competitive
https://www.dummies.com/programming/cloud-computing/hybrid-cloud/the-b 62
usiness-benefits-of-paas-in-cloud-computing
/
ADVANTAGES OF INFRASTRUCTURE AS A
SERVICE??

63
IAAS-BENEFITS

• Business Agility
Business agility is the ability to provision and de-provision
resources in lesser time, enabling the infrastructure size to
match demand. For example, it can be used to support an e-
commerce website which is in need of reacting to various
types of demand loads every day. It can also be used to
support the test and development that occurs on a non-
predictable basis.
64
IAAS-BENEFITS

• Cost Reduction
Cost reduction can be done in multiple ways. It can be done by provisioning
workloads on an as-needed basis, eliminating the need to purchase or lease
hardware, or automatically provisioning them to eliminate manual intervention
and labor costs. Cost can also be reduced by moving from capex to an opex
model, providing a consumption based approach to IT rather than paying off IT
hardware over a three-year period with greater chance of obsolescence during
the term. For example, it can be used to support test and development cycles
that are measured in weeks, or to support e-commerce websites for shopping
periods or online gaming.
65
IAAS-BENEFITS

• Local Presence
Local presence is the ability to provision infrastructure in a
geographic presence where clients do not have IT operations
and to achieve compliance with local country regulations in
respect to data sovereignty. For example, to provision IT
infrastructure in new regions.

66
IAAS-BENEFITS

• Rigid Elasticity
It can be said to be the ability to automatically scale due to
unexpected events. Every enterprise has some kind of disaster
recovery plan, but the technology behind them is usually very
expensive. IaaS provides a consolidated disaster recovery
infrastructure, reducing costs and increasing manageability.

67
IAAS-BENEFITS

• Focus on business growth

Time, money and energy spent making technology decisions
and hiring staff to manage and maintain the technology
infrastructure is the time which is not spent on business
growth. When the infrastructure is moved to a service based
model, organizations are able to focus their time and resources
where they belong, on developing innovations in applications
and solutions.
68
 CHALLENGES OF SERVICES CLOUD
COMPUTING

• Cost
• Service Provider Reliability
• Downtime
• Password Security
• Data privacy
• Vendor lock-in
https://www.techwell.com/techwell-insights/2017/10/6-major-challenges-cloud-computing70
MARKET-ORIENTED ARCHITECTURE

• NEED:
– Consumers will require different QoS to be maintained by
their providers.
– Providers will need to consider and meet different QoS
parameter of each individual consumer
– So market oriented resource management is necessary to
regulate the supply and demand cloud resources at market
equilibrium. 71
MARKET ORIENTED CLOUD ARCHITECTURE

72
MARKET ORIENTED CLOUD
ARCHITECTURE

• Users/Brokers :
Submit service request from any where in the world.
• SLA Resource Allocator :
Interface between cloud provider and users.

73
SLA RESOURCE ALLOCATOR

74
EMERGING CLOUD PLATFORMS

* Amazon EC2
• Elastic Compute Cloud (EC2)
• Amazon Machine Image (AMI)
• EC2 Uses XEN Virtual Machine
• Virtual Os :Linux, Solaris, Windows
• Simple Storage Service• Elastic IP address
• Amazon Cloud Watch• Realiability
75
EMERGING CLOUD PLATFORMS

* Google App Engine:

• For developing and hosting web application in google
managed datacenter
• Automatic scaling
• Free up to certain level of consumption
• Support for python, java, and Go
76
EMERGING CLOUD PLATFORMS

* Microsoft Live Mesh

• Access stuffs on your computer from almost anywhere
• Free to use internet based file synchronization application
• Remote desktop access via internet

77
EMERGING CLOUD PLATFORMS

* Sun Grid
• Now Know as oracle Grid
• Open source batch queuing system
• Used on computer farm or high performance computing
cluster

78
LIMITATIONS OF PRESENT SERVICE
PROVIDERS

79
GLOBAL CLOUD EXCHANGE AND
MARKETS

Limitations of present service providers:

• Inflexible pricing
• Consumers are restricted to offering from a single provider at
a time
• Unable to swap one provider for another
• No standard interface
80
GLOBAL CLOUD EXCHANGE

Global Cloud Exchange:

• Market directory
• Banking system
• Brokers
• Price setting mechanism
• Admission control mechanism
• Resource management system
• Consumers utility function
81
• Resource management proxy
CHALLENGES

• Unwillingness to shift from traditional controlled environment

• Regulatory pressure
• How to obtain restitution in case of SLA violation

• https://
www.slideshare.net/parakkajithin/market-oriented-cloud-com
82
puting
SIMULATION TOOLKITS

• Cloud Simulation Frameworks

- CloudSim:
- GridSim:

83
CLOUDSIM: A COULD SIMULATION
FRAMEWORK

• Introduction
• Features
• Architecture
• Implementing the Cloud

84
 INTRODUCTION CLOUDSIM

• CloudSim is an open source software under the GPL license

(GENERAL PUBLIC LICENSE- http://
www.gnu.org/copyleft/gpl.html ) developed in the Cloud
Computing and Distributed Systems (CLOUDS) Laboratory, at
the Computer Science and Software Engineering Department
of the University of Melbourne
• http://code.google.com/p/ cloudsim /
85
 INTRODUCTION CLOUDSIM

CloudSim simulation framework enables users to:

• Test application in controlled and repeatable environment
• Find the system bottlenecks without the need of real clouds
• Try different configurations for developing adaptive
provisioning techniques

86
FEATURES OF CLOUDSIM

CloudSim features include :

• Modeling and simulating large scale data centers
• Modeling and simulating virtualized server hosts
• Modeling and simulating energy-aware computational resources
• Modeling and simulating federated clouds
• Inserting simulation elements dynamically
• User-defined policies for allocation of hosts to virtual machines
87
CLOUDSIM ARCHITECTURE

• CloudSim model of Cloud Computing architecture consists of

three layers the system layer, the core middleware, and the
user-level middleware as shown in figure below.
• These three layers corresponding to three top layers of cloud
computing architecture Iaas, Paas and SaaS respectively.

88
CLOUDSIM ARCHITECTURE

89
IMPLEMENTING THE CLOUD

• CloudSim is implemented using existing simulation libraries and

frameworks such as GridSim and SimJava in order to handle the low-
level requirements of the system. In particular, the components of
handling events and message passing in SimJava can be reused in
CloudSim. Similarly, using GridSim simplifies the implementation of
networking, information services, files, users, etc. We now discuss
main components of a cloud computing infrastructure which are
implemented in the simulator CloudSim. Figure below shows complete
components of CloudSim.
90
CLOUDSIM COMPONENTS

91
COMPLETE COMPONENTS OF CLOUDSIM

• Data center:
In CloudSim, data center is used to model the core services at
the system level of a cloud infrastructure. It consists of a set of
hosts which manage a set of virtual machines whose tasks are
to handle "low level" processing, and at least one data center
must be created to start the simulation.

92
COMPLETE COMPONENTS OF CLOUDSIM

• Host:
This component is used to assign processing capabilities
(which is specified in the milion of instruction per second that
the processor could perform), memory and a scheduling policy
to allocate different processing cores to multiple virtual
machines that is in the list of virtual machines managed by the
host.

93
COMPLETE COMPONENTS OF CLOUDSIM

• Virtual machine:
This component manages the allocation of different virtual
machines different hosts, so that processing cores can be
scheduled (by the host) to virtual machines. This configuration
depends on particular application, and the default policy of the
allocation of virtual machines is "first-come, first-serve"

94
COMPLETE COMPONENTS OF CLOUDSIM

• Datacenter broker:
The responsibility of a broker is to meditate between users and service
providers, depending on the requirement of quality of service that the
user specifies. In other words, the broker will identify which service
provider is suitable for the user based on the information it has from the
Cloud Information Service, and negotiates with the providers about the
resources that meet the requirement of the user. The user of CloudSim
needs to extend this class in order to specify requirement in their
experiments.
95
COMPLETE COMPONENTS OF CLOUDSIM

• Cloudlet:
This component represents the application service whose complexity
is modeled in CloudSim in terms of the computational requirements.
• CloudCoordinator:
This component manages the communication between other
CloudCoordinator services and brokers, and also monitor the internal
state of a data center which will be done periodically in terms of the
simulation time.
96
TÌM HIỂU THÊM VỀ CÁC CÔNG CỤ MÔ
PHỎNG CLOUD KHÁC NHƯ:

• GRIDSIM ?
• SIMJAVA??
• …………….. ??

97
98
NOTE!

• Incorporating Grid resource failures during runtime

• supporting advance reservation of a grid system
• reading workload traces from supercomputers to simulate a realistic grid
environment
• supporting data grid simulations
• utilizing a network topology to link together resources and other entities
• simulating background network traffic based on a probabilistic distribution to
simulate a congested public network
• simulating an experiment with multiple virtual organizations
99
http://cloud-simulation-frameworks.wikispaces.asu.edu/
Q&A

100