Cloud Computing

Concept, Technology & Architecture

Chapter 05
Cloud-Enabling Technology
 Contents
2

 Contemporary technologies that realize modern-

day cloud computing platforms and innovations
are discussed, particularly data centers,
virtualization, and Web-based technologies.
 5.1 Broadband Networks and Internet Architecture
 5.2 Data Center Technology
 5.3 Virtualization Technology
 5.4 Web Technology
 5.5 Multitenant Technology
 5.6 Service Technology
 5.7 Case Study Example
 Broadband Networks and Internet
Architecture

• All the clouds must be connected to a network

• Inherent dependency on internetworking

• Private clouds have dedicated network links in LANs

• Most clouds are internet-enabled

 Internet Service Provider(ISPs)

• Deploy the largest backbone networks

• ISPs can freely deploy, operate and Mange

• Interconnect with other IPSs

• Governmental and regulatory laws dictate the service provisioning conditions

• ISPs mainly classified into three tiers
International ISPs:  AT&T WorldNet, IBM Global Network, MCI, Netcom, UUNet, and PSINet
National ISPs:JIO, BSNL, IDEA, AIRTEL
Local ISPs: SKY-NET, SYSLINK, Excell broadband
Figure 5.1
4

 Figure 5.1 - Two

messages travel over
dynamic network routes
in this ISP
internetworking
configuration.
 Figure 5.2
5

 Figure 5.2 - An
abstraction of the
internetworking structure
of the Internet.
Fundamental Components of Internetworkin
g
Connectionless Packet Switching (Datagram Netwo
rks)

• End to end data flows are divided into packets

• Store and forward

• Each packet has IP and MAC address

Router-Based Interconnectivity

• Router device is connected to multiple networks

• Maintains the network topology information

• provides alternative paths

 Figure 5.3
6

 Figure 5.3 - Packets

traveling through the
Internet are directed
by a router that
arranges them into a
message.
 Figure 5.4
7

 Figure 5.4 - A generic view of the Internet reference model and

protocol stack.
Technical and Business Considerations

•Connectivity issues
•Network bandwidth and latency issues
•Cloud carrier and cloud provider selection

S –Erouter- 4mbps
E-lISP- 10mbps
lISP-NISP-200mbps
NISP-Dc-100mbps
Dc-Ehost- 5mbps
 Figure 5.5
8

 Figure 5.5 - The internetworking architecture of a private cloud. The

physical IT resources that constitute the cloud are located and managed
within the organization.
The internetworking architecture over th
e internet
On-Premise IT Resources Cloud-Based IT Resources

internal end-user devices access internal end-user devices access

corporate IT services through the corporate IT services through an
corporate network Internet connection

internal users access corporate IT internal users access corporate IT

services through the corporate Internet services while roaming in external
connection while roaming in external networks through the cloud provider’s
networks Internet connection

external users access corporate external users access corporate IT

IT services through the services through the cloud provider’s
corporate Internet connection
Internet connection
 Summary of Broadband Networks and
Internet
11
 Cloud consumers and cloud providers typically use
the Internet to communicate, which is based on a
decentralized provisioning and management model
and is not controlled by any centralized entities.
 The main components of internetworking
architecture are connectionless packet switching and
router-based interconnectivity , which use network
routers and switches.
 Networks bandwidth and latency are characteristics
that influences QoS, which is heavily impacted by
network congestion.
 5.2 Data Center Technology
(1/2)
 Grouping IT resources in 12close proximity with one
another allows for power saving, higher efficiency in
sharing resources, and improve accessibility for IT
personnel. Following issues are concerned:

1. Virtualization
2. Standardization and Modularity
3. Automation (self-configuration, recovery)
4. Remote Operation and Management
5. High Availability (through redundancy)
 5.2 Data Center Technology
(2/2)
6. Security-Aware Design, Operation,
13 and
Management (outsourcing resources)
7. Facilities
8. Computing Hardware (standardized commodity servers)
9. Storage Hardware (array, hot-swapping, storage
virtualization, fast data replication, SAN, NAS,…)
10. Network Hardware
 Virtualization
• Data center consist of both physical and virtua
lized IT resources.

• All the physical components are grouped into

single network

• The resource abstraction and control on virtua

lization layer
Figure 5.7
14

 Figure 5.7 - The common

components of a data
center working together
to provide virtualized IT
resources supported by
physical IT resources.
 Standardization and Modularity
• Commodity Hardware with modular architectu
re, aggregating multiple identical building bloc
ks of infrastructure
• Key requirements for reducing investment and
operational cost
 Automation
• Specialized platforms and tools are needed

• Resource Provisioning, configuration, patching

and monitoring

• Self-configuration and self-recovery

 Remote Operation and Management

• Used for most of operational and administrativ

e tasks of IT resources

• Equipment handling and cabling or hardware-l

evel installation and maintenance not controll
ed from remotely.
 High Availability
• Data center Outage significantly impacts IT Ser
vices availability
• Higher levels of redundancy to sustain availabi
lity
Security-Aware, operation and Management

• Ubiquitous access, on-demand provisioning ra

pid elasticity and pay-per-use.
• Physical and logical access controls and data re
covery strategies.
 Facilities
• Computing, Storage and network equipme
nt
• power, cabling, cooling, fire protection
 Computing Hardware
•Supports Different processing architectures like x86-32,x86-64 and RISC
•Power efficient multi-core CPU
•Redundant and hot-swappable components

Blade Server
Rack Server
 STORAGE HARDWARE
• Hard Disk Array: RAID
• I/O caching
• Storage Virtualization
• Fast Data Replication Mechanisms

Network Storage device

• storage area network(SAN)
• network attached storage(NAS)
NAS vs SAN
NAS Benefits
• Relatively inexpensive
• 24/7 and remote data availability
• Salability
• Redundant storage architecture
• Automatic backups to other devices and cloud
• Flexibility
SAN Benefits
• Extremely fast data access
• Dedicated network for storage relieves stress
on LAN
• Highly expandable
• OS level (block level) access to files
• High quality-of-service for demanding applicati
ons such as video editing
 Network Hardware
• Carrier and External Networks Interconnection
• Web-Tier Load Balancing and Acceleration
• LAN Fabric
• SAN fabric
• NAS Gateways
Summary Data Center Technology
(1/2)
 A data center is a specialized
15
IT infrastructure that
houses centralized IT resources, such as servers,
databases, and software systems.
 Data center IT hardware is typically comprises of
standardized commodity servers of increased
computing power and storage capacity, while storage
system technologies include disk arrays and storage
virtualization. Technologies used to increase storage
capacity include DAS, SAN, and NAS.
Summary of Data Center Technology (2/2)
16

 Computing hardware technologies include rack-

mounted server arrays and multi-core CPU
architectures, while specialized high-capacity
network hardware and technology, such as content-
aware routing, LAN and SAN fabrics, and NAS
gateways, are used to improve network connectivity.
 5.3 Virtualization Technology
(1/2)
 Most types of IT resources
17
can be virtualized:
servers, storage, network, power.

 Resource Replication
 Virtual disk images can be accessible using simple file
operations, such as copy, move, and paste by the host’s OS for
replication purpose
VM Architectures
VM Operations
 Server Consolidation
Different virtual servers share one physical server
 5.3 Virtualization Technology
(2/2)
Operating System-Based 18Virtualization
Hardware-Based Virtualization
 Virtualization Management
 Virtualization infrastructure management ( VIM) collectively
manage virtual IT resources running on a centralized module.
 Other Considerations
 Performance overhead

 Special hardware compatibility Portability

 (Open Virtualization Format OVF)
 Operating System-Based Virtualization

• Installation of virtualization software in pre-exi

sting operating system(Host)

• Host operating system responsible to provide

hardware device access

• Create more Virtual Servers

Figure 5.8
19

 Figure 5.8 - The different

logical layers of
operating system-based
virtualization, in which
the VM is first installed
into a full host operating
system and subsequently
used to generate virtual
machines.
 Cont..
• Hardware independence is enabled

Host operating system provide the following ser

vices
• Backup and Recovery
• Integration to Directory Services
• Security Management
 Demands and issues
• Host operating system consumes CPU, Memor
y and other hardware resources

• Virtual operating system instructions have to p

rocess several layers

• Host operating requires individual licence

 Hardware-Based Virtualization
• Bypass the host operating system

• With out host operating interaction, it allows virtual

servers to interact with hardware

• VMM require a negligible amount of storage

• VMM is a thin layer software

 Figure
5.9
20

 Figure 5.9 - The different logical layers of hardware-based

virtualization, which does not require another host operating system.
• Device drivers and system services are optimiz
ed

Issues of Hardware-based Virtualization

• Compatibility with hardware devices
• Device drivers may not be available
 Virtualization Management
• Many administrative task can be performed more
easily

• Modern virtualization software several advanced

management functions that can automate admini
stration tasks

• Virtualization infrastructure Management (VIM) to

ols that collectively manage virtual resources
 Other considerations
• Performance overhead

• Special hardware compatibility

• Portability
 Best virtualization software - at a glance

• VMware Workstation Player

• VirtualBox
• Parallels Desktop
• QEMU
• Citrix Hypervisor
• Xen Project
• Microsoft Hyper-V
 Case study 1: 
Financial services company running out of s
pace
As computing needs continued to grow, an online financial serv
ices company faced a problem all too common in today’s dat
a centers. The company was running out of space to house p
hysical servers, and its existing cooling infrastructure couldn't
keep up. New workloads meant the addition of physical syste
ms, and valuable time was spent configuring those systems a
nd balancing power distribution. Finally, the company turned
to server virtualization technology to help solve its space pro
blems and improve efficiency. Today, the company is 75% virt
ualized and runs 200 VMs on just 10 physical servers, leading
to an estimated 33% savings in power use.
 Summary of Virtualization Technology
21

 Server virtualization is the process of abstracting IT

hardware into virtual servers using virtualization
software.
 Virtualization provides hardware independence,
server consolidation, and resource replication, and
further supports resource pooling and elastic
scalability.
 Virtual servers are realized through either operating
system-based or hardware-based virtualization.
Web Technology

computers communicate with each other using

markup languages and multimedia packages
 Communication protocols
• HTTP
• SMTP
• IP
 5.4 Web
Technology
 Basic Web Technology 22
 Uniform Resource Locator (URL)

 Hypertext Transfer Protocol (HTTP)

 Markup Language (HTML, XML)

 Web Applications (Figure 5.10)

 Presentation layer (user interface)

 Application layer (application logic in application server)

 Data layer (data store in data server)
 Figure
5.10
23

 Figure 5.10 - The three basic architectural tiers of Web applications.

 Summary of Web
Service
 Web technology is very commonly
24
for cloud service
implementations and for front-ends used to remotely
manage cloud-based IT resources.
 For instance, typical PaaS offerings have separate instances of
the Web server, application server, and data server.
 Fundamental technologies of Web architecture
include the URL, HTTP, HTML, and XML.
 5.5 Multitenant
Technology
 Multitenant – a single instance
25
of an IT resource
serves multiple consumers (tenants).
 Multitenant application architecture is often
significantly more complex than that of single-tenant
applications.
 Multi-tenant applications need to support the
sharing of various artifacts by multiple users
(including portals, data schemas, middleware, and
databases), while maintaining security levels that
segregate individual tenant operational
environments.
 5.5 Characteristics of
Multitenant
 Usage isolation 26

 Data security
 Recovery
 Application upgrades
 Scalability
 Metered usage
 Data tier isolation

 Multitenancy is sometimes mistaken for virtualization

because the concept of multiple tenants is similar to the
concept of virtualized instances.
 Customize features
• User Interface
• Business process
• Data Model
• Access Control
 Figure 5.11
27

Figure
 5.11 - A multitenant application
that is serving, multiple cloud service,
consumers simultaneously
.
 Multitenancy VS. Virtualization
28

 The differences lie in what is multiplied within a

physical server acting as a host:
 With virtualization:
 Multiple virtual copies of the server environment can be
hosted by a single physical serve. Each copy can be provided
to different users, can be configured independently, and can
contain its own operating system and applications.
 With multitenancy:
 A physical or virtual server hosting an application is
designed to allow usage by multiple different users. Each
user feels as though they have exclusive usage of the
application.
 5.6 Service
Technology
 Along with XML, the core29technologies behind Web
services are represented by the following industry
standards:
 Web Service Description Language (WSDL)
 XML Schema Definition Language (XML Schema)
 SOAP (Simple Object Access Protocol)
 Universal Description, Discovery, and Integration (UDDI)
 Note that: These 4 technologies collectively form the
first generation of Web service technology. The 2nd
generation (WS-*) addresses additional features, such
as security, reliability, transactions, routing, and
business process automation.
 WSDL Document

Element Description

<types> Defines the (XML Schema) data types used by the web
service
<message> Defines the data elements for each operation

<portType> Describes the operations that can be performed and the

messages involved.

<binding> Defines the protocol and data format for each port type
 WSDL Operations
One-way: The operation can receive a m
essage but will not return a response

Request-response:The operation can rec

eive a request and will return a respons
e

Solicit-response:The operation can send

a request and will wait for a response

Notification: The operation can send a

message but will not wait for a response
 SOAP
• SOAP – Simple Object Access Protocol.

• SOAP relies heavily on XML, and together with schemas, defines a very str
ongly typed messaging framework.

• Every operation the service provides is explicitly defined, along with the X
ML structure of the request and response for that operation.

• Each input parameter is similarly defined and bound to a type: for exampl
e an integer, a string, or some other complex object.

• All of this is codified in the WSDL – Web Service Description (or Definition,
in later versions) Language.
 WSDL Binding to SOAP
• <message name="getTermRequest">
  <part name="term" type="xs:string"/>
</message>

<message name="getTermResponse">
  <part name="value" type="xs:string"/>
</message>

<portType name="glossaryTerms">
  <operation name="getTerm">
    <input message="getTermRequest"/>
    <output message="getTermResponse"/>
  </operation>
</portType>

<binding type="glossaryTerms" name="b1">
   <soap:binding style="document"
   transport="http://schemas.xmlsoap.org/soap/http" />
   <operation>
     <soap:operation soapAction="http://example.com/getTerm"/>
     <input><soap:body use="literal"/></input>
     <output><soap:body use="literal"/></output>
  </operation>
</binding>
 Figure
5.12
30

 Figure 5.12 - An overview

of how first-generation
Web service technologies
commonly relate to each
other
 REST Service and Service Agents
31

 REST services are designed according to a set of

constraints that shape the service architecture to
emulate the properties of the WWW.
 Service agents are event-driven programs designed
to intercept messages at runtime, either active agent
or passive agent.
 Falling under the umbrella of service technology is the
large of middle platform. Two main categories are the
enterprise service bus (ESB) and the orchestration
platform.
 Summary of Service Technology
32

 Web-based services such as Web services and REST

services rely on non-proprietary communications
and technical interface definitions to establish
standardized communications frameworks based on
Web technology.
 Service agents provide event-driven
runtime processing that can be applied to
numerous functional areas within clouds.
 Service middleware, such as ESBs and orchestration
platforms, can be deployed on clouds.
 DTGOV Case Study

DTGOV has assembled cloud-aware infrastructures in each of it

s data centers, which are comprised of the following compon
ents:

• Tier-3 facility infrastructure, which provides redundant config

urations for all of the central subsystems in the data center f
acility layer.

• Redundant connections with utility service providers that hav

e installed local capacity for power generation and water sup
ply that activates in the event of general failure.
 Cont..
• An internetwork that supplies an ultra-high bandwidth i
nterconnection between the three data centers through
dedicated links.

• Redundant Internet connections in each data center to

multiple ISPs and the .GOV extranet, which interconnect
s DTGOV with its main government clients.

• Standardized hardware of higher aggregated capacity tha

t is abstracted by a cloud-aware virtualization platform.
 Cont..
• Physical servers are organized on server racks, each of which h
as two redundant top-of-rack router switches (layer 3) that ar
e connected to each physical server.

• These router switches are interconnected to LAN core-switche

s that have been configured as a cluster.

• The core-switches connect to routers that supply internetwor

king capabilities and firewalls that provide network access con
trol capabilities.
 Cont..
• A view of the server network connections inside the DTGOV d
ata center.

• A view of the storage system network connections inside the

DTGOV data center.

• A view of how two data center are connected each other in D

TGOV
 Figure 5.13
33

 Figure 5.13 - A view of

the server network
connections inside the
DTGOV data center.
 Figure 5.14
34

 Figure 5.14 - A view of the

storage system network
connections inside the
DTGOV data center.
 Greenlight Technology Simplified
Server Consolidation

What is Server Consolidation?

Server consolidation is a strategy to reduce the total number of physic

al servers or server locations that an organisation requires. The practice
has been developed in response to the problem of server sprawl, a situ
ation in which multiple, underutilized servers take up more space and c
onsume more resources than can be justified by their workload. Reduci
ng the number of servers brings down the associated costs with servers
and compute resources for an organisation. The more dense the consoli
dation, the higher the ROI is for the migration and the sooner the cost b
enefit of consolidation is realised.
 Case Study
• Our client is a software development company in the application of
handheld devices and industrial inventory management systems for
large freight companies, specialist firms, and warehouses. Their exis
ting IT systems had some virtualisation in place, along with single pu
rpose server blades and standalone servers for other organisation n
eeds.
• The systems had been over provisioned and the resources were und
erutilised. Although virtualisation was already in place, they were us
ing SAN storage with limited extensibility and all the existing hardwa
re was out of warranty and in excess of 4 years in age. The only cost
effective solution was to migrate the environment to new hardware
with much greater performance, as well as increased data density.
 Initial Requirement
• Our client had previously made a large invest
ment in hardware to run their operation; this
hardware was aging and becoming unreliable.
With current generation hardware, it was strai
ghtforward to have a similar level of resources
on new infrastructure.
 Existing hardware
• The existing hardware was a complex mix of physical machines,
server blades, and virtual machines that were VMware based.

• In line with the customers objectives for the virtualization proj

ect the decision was made to take the entire environment and c
onvert it to VMware virtual machines running on ESXi 5.

• The existing environment consisted of two racks of equipment’s

containing a fully populated HP Blade Centre, a selection of old
er generation HP servers and older generation SAN.
 New layout and hardware

• By leveraging the performance and memory

density of the latest generation of Dell rack m
ount servers we are able to condense the enti
re environment on to a single two rack unit vi
rtualisation host, which was relocated into Gl
obal Switch and a single two rack unit storage
server for local file access located in the offic
e.

• The new Dell server has redundant power su

pplies, fault tolerant RAID arrays and accelera
ted cache for optimal speed and redundancy.

• We also redeployed one of the existing HP se

rvers to provide local replication of the file st
orage via DFS. The previously slow and unsta
ble VPN connections that were coming into t
he office for remote workers are now being t
erminated on the Data Centre side on a Cisco
ASA with SSL VPN’s with significantly increase
d bandwidth and stability.
 The Conversion Process
• We leveraged the features in VMware converter to
convert the existing virtual machines and physical
machines to VMware virtual machines.

• The conversion was done in a staggered fashion o

utside of office hours to avoid interrupting busines
s operations.

• The VMware converter allows for machines to be

processed in batches, with a technician working o
n Virtual Machine configuration settings such as n
etworking, then reconfiguring each machines afte
r the move. The migrations are still ongoing in the
background minimising the time wastage of techni
cians sitting around waiting for the conversions.
RESULT
Cont..
 Conclusion
• With the virtualization project now complete,
our client is now reaping the benefits of the co
nsolidated infrastructure. Remote workers no
w no longer suffer from bandwidth issues and,
applications hosted on the new infrastructure
now run much faster.