Y. V.

Lokeswari - AP/ CSE

SSN College of Engineering

Reference: Distributed and Cloud Computing

K. Hwang, G. Fox and J. Dongarra
 System Models for Distributed and Cloud Computing
 Cluster and co-operative computers
 Grid computing Infrastructures
 Peer –to-Peer Network
 Cloud Computing over Internet.
 3 Cloud Service Models
 Cloud Computing Challenges
 IoT
 Service Oriented Architecture (SOA)
 Performance parameters
 System Attacks
 Distributed and cloud computing systems are built
over large number of computer nodes
 These node machines are interconnected by SANs,
LANs, or WANs.
 One can build a massive systems with millions of
computers connected to network.
 Massive systems are considered highly scalable.
 Massive systems are classified as clusters, P2P
networks, computational grids and internet of clouds.
 Computing cluster consists of inter-connected stand-
alone computers
 They work cooperatively as a single integrated resource.
 They produce impressive results in handling heavy
workloads with large data sets.
A Typical Cluster Architecture

Copyright © 2012, Elsevier Inc. All rights reserved. 1-6

 Single System Image (SSI)
 Gref Pfister [2] : An ideal cluster should merge multiple
system images into a SSI.
 Cluster Designers desire a Cluster Operating System or
some middleware to support SSI at sharing of CPU,
Memory (DSM) and IO devices.
 SSI makes a cluster to appear like a single machine to a
user.
 A cluster with multiple images is a collection of
independent computers.
 Most of clusters have Linux OS.
 Challenge: Unfortunately, a cluster-wide OS for complete
resource sharing is not yet available. Middleware is
required. Otherwise cluster nodes cannot work together to
produce cooperative results.
 Internet services enables local computers to connect to
remote computes.
 Telnet commands.
 HTTP web service.
 Grid allows close interaction among applications
running on distant computers simultaneously.
 Forbes Magazine : Global growth of IT-based
economy from $1 trillion in 2001 to $20 trillion in 2015
• A computing grid offers an infrastructure that couples
computers, software, middleware, special instruments,
and people and sensor together, which is usually
constructed across LAN, WAN, or Internet backbone.
• 2 categories: computational/data grids and P2P grids.
Computational or data grid providing computational utility
, data and information services through resource sharing
cooperation among participating organizations
Courtesy of Z. Xu, Chinese Academy of science, 32004
A Typical Computational Grid

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 13

 A distributed system architecture
 Each computer in the network can act as a client or server
for other network computers.
 No centralized control or No central database.
 Typically many nodes, but unreliable and heterogeneous
 Physical network is a ad-hoc network formed using TCP/IP.
 Take advantage of distributed, shared resources (bandwidth,
CPU, storage) on peer-nodes
 Fault-tolerant, self-organizing
 Operate in dynamic environment, frequent join and leave
the system freely. (No Master- Slave relationship among
peers)
Overlay network - computer network built on top of another network.
• Nodes in the overlay can be thought of as being connected by virtual or
logical links, each of which corresponds to a path, perhaps through many
physical links, in the underlying network. (Unstructured and structured
Overlays)
• For example, distributed systems such as cloud computing, peer-to-peer
networks, and client-server applications are overlay networks because their
nodes run on top of the Internet.
P2P Design Challenges : Hardware, software and network
Too many hardware models and architectures to select, incompatibility
between software, OS and different network connection protocols.
Performance affected by routing efficiency and self-organizing nature of
peers. Management is difficult, lacks security, No fault-tolerance, load
balancing, Not virus free.
 Historical roots in today’s
Internet apps
 Search, email, social networks
 File storage (Live Mesh, Mobile
Me, Flicker, …)
 Computational science is changing to be data-
intensive.
 Working with large data sets will mean sending the
computations (programs) to the data, rather than
copying the data to the workstations.
 A cloud infrastructure provides a framework to
manage scalable, reliable, on-demand access to
applications
 A cloud is the “invisible” backend to many of
our mobile applications
 A model of computation and data storage based
on “pay as you go” access to “unlimited”
remote data center capabilities Copyright © 2012, Elsevier Inc. All rights reserved.
 Basic Concept of Internet Clouds
• Cloud computing is the use of computing resources (hardware and
software) that are delivered as a service over a network (typically the
Internet).
• A cloud is a pool of virtualized computer resources, which can
host a variety of different workloads, including batch and
interactive applications
• Cloud ecosystem must be designed to be secure, trustworthy and
dependable.

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 20

  Classical Computing  Cloud Computing
 Buy & Own  Subscribe
 Hardware, System  Use

Software,
Every 18 months?

Applications often to
meet peak needs.
 Install, Configure, Test,
Verify, Evaluate
 Manage
 ..  $ - pay for what you use,
 Finally, use it
based on QoS

 $$$$....$(High CapEx)
(Courtesy of Raj Buyya, 2012)
Copyright © 2012, Elsevier Inc. All rights reserved.
Infrastructure as a service (IaaS)
 Most basic cloud service model
 Cloud providers offer computers, servers, storage, network and
data centers as physical or more often as virtual machines.
 Virtual machines are run as guests by a hypervisor, such as Xen or
KVM.
 Cloud users deploy their applications by then installing operating
system images on the machines as well as their application
software.
 Cloud providers typically bill IaaS services on a utility computing
basis, that is, cost will reflect the amount of resources allocated
and consumed.
 Examples of IaaS include: Amazon CloudFormation (and
underlying services such as Amazon EC2), Rackspace Cloud,
Terremark, and Google Compute Engine.
Platform as a service (PaaS)
 Cloud providers deliver a computing platform typically
including operating system, programming language
execution environment, database, and web server.

 Application developers develop and run their software on a

cloud platform without the cost and complexity of buying
and managing the underlying hardware and software layers.

 Examples of PaaS include: Amazon Elastic Beanstalk, Cloud

Foundry, Heroku, [Link], EngineYard, Mendix, Google
App Engine, Microsoft Azure and OrangeScape.
Software as a service (SaaS)
 Cloud providers install and operate application software in
the cloud and cloud users access the software from cloud
clients.
 The pricing model for SaaS applications is typically a
monthly or yearly flat fee per user, so price is scalable and
adjustable if users are added or removed at any point.
 Examples of SaaS include: Google Apps, innkeypos,
Quickbooks Online, Limelight Video Platform,
[Link], and Microsoft Office 365.
 Scalability

Reliability

Billing

Utility & Risk

Management

Software Eng.
Complexity
Programming Env.
& Application Dev.

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 29

 SOA is an evolution of distributed computing based on the
request/reply design paradigm for synchronous and
asynchronous applications.
 An application's business logic or individual functions are
modularized and presented as services for consumer/client
applications.
 Key to these services - their loosely coupled nature;
 i.e., the service interface is independent of the implementation.
 Application developers or system integrators can build
applications by composing one or more services without
knowing the services' underlying implementations.
 For example, a service can be implemented either in .Net or J2EE, and
the application consuming the service can be on a different platform
or language.
 SOA services have self-describing interfaces in platform-independent XML
documents.
 Web Services Description Language (WSDL) is the standard used to describe the
services.
 SOA services communicate with messages formally defined via XML Schema
(also called XSD).
 Communication among consumers and providers or services typically happens in
heterogeneous environments, with little or no knowledge about the provider.
 Messages between services can be viewed as key business documents processed in an
enterprise.
 SOA services are maintained in the enterprise by a registry that acts as a
directory listing.
 Applications can look up the services in the registry and invoke the service.
 Universal Description, Definition, and Integration (UDDI) is the standard used
for service registry.
 Each SOA service has a quality of service (QoS) associated with it.
 Some of the key QoS elements are security requirements, such as authentication and
authorization, reliable messaging, and policies regarding who can invoke
services.
 Smart
Earth:
An
Internet of
Things IBM
Dream
Smart Earth

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 34

 (courtesy of Wikipedia, 2010)

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 35

Transparent Cloud Computing Environment
Parallel and Distributed Programming

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 37

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 38
 Size – increasing performance by increasing machine
size
 Software – upgrade to OS, libraries, new apps.
 Application – matching problem size with machine size
 Technology – adapting system to new technologies
System Scalability vs. OS Multiplicity

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 40

System Availability vs. Configuration Size :

Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 41

Security: System Attacks and Network Threads

[Link]
[Link]
Copyright © 2012, Elsevier Inc. All rights reserved. 1 - 43
 How Cluster differs from Grid
 What is Overlay Network in P2P network
 Differentiate Grid & Cloud
 Motivation for Cloud Computing
 Services in Cloud with example systems
 Security Issues in Cloud
 Cluster and co-operative computers
 Grid computing Infrastructures
 Peer –to-Peer Network
 Cloud Computing over Internet.
 3 Cloud Service Models
 Cloud Computing Challenges
 IoT
 Service Oriented Architecture (SOA)
 Performance parameters
 System Attacks
1. K. Hwang, G. Fox, and J. Dongarra, Distributed and Cloud
Computing: from Parallel Processing to the Internet of Things
Morgan Kauffmann Publishers, 2011
2. Greg F Pfister, In search of clusters, Second ed., Prentice hall , 2001

3. R. Buyya, J. Broberg, and A. Goscinski (eds), Cloud Computing:

Principles and Paradigms, ISBN-13: 978-0470887998, Wiley Press,
USA, February 2011.
4. T. Chou, Introduction to Cloud Computing: Business and
Technology, Lecture Notes at Stanford University and at Tsinghua
University, Active Book Press, 2010.
5. T. Hey, Tansley and Tolle (Editors), The Fourth Paradigm : Data-
Intensive Scientific Discovery, Microsoft Research, 2009.