Presented By: Veena.K.P Mca S5 Roll No:28

Presented by:
Veena.K.P
MCA S5
Roll No:28
Topics covered
• Introduction
• Definition
• Why clusters?
• Benefits of clusters
• Types of clusters
• Architecture
• Issues to be considered
• Future trends
INTRODUCTION
Definition
Cluster Computing is the mechanism of

performing computational task using
clusters
What is a Cluster?
A cluster is a collection of connected,

independent computers that work
together to solve a problem.
A Typical Cluster
• Many standalone computers (nodes).

• All of the cluster can work together on a
single problem at the same time.
• Portions of the cluster can be working on
different problems at the same time.
• Connected together by a network.
• Administered as a single “machine”.
Why Clusters ?
• High cost of ‘traditional’ High Performance

Computing
• Increased need for High
Performance Computing
• Best price and performance ratio
• In areas like data mining to fault tolerant
performance clustering technology has
become important.
Benefits of clusters
• Reduced Cost
• Scalability
• Availability
• Easy management
• Reliability
Types of clusters
A. Based on their use:
• High availability or fail over cluster
• Load balancing cluster
• Parallel processing clusters
Grid Computing
B. Based on Architecture:
• Close cluster
• Open cluster
High availability or Failover cluster
 provides uninterrupted availability of data

 used for mail, file and print, web or
application servers.
High availability cluster.
Load balancing cluster
 distributes incoming requests for resources

among multiple nodes.
 increases reliability and availability.
Load balancing cluster
Parallel processing clusters
 multiple processors share a single memory

and bus interfaces within a single computer.
 used in mathematical computations,
scientific analysis etc
Close clusters
They hide most of the cluster behind the

gateway node. Consequently they need less
IP addresses and provide better security.
They are good for computing tasks.
Close clusters
High Speed Network
compute compute compute compute File

node node node node Server
node
Service Network
gateway Front-end
node
External Network
Open clusters
All nodes can be seen from outside,and

hence they need more IPs, and cause more s
ecurity concern .But they are more flexible
and are used for internet/web/information se
rver task.
Open clusters
High Speed Network
compute compute compute compute File

node node node node Server
node
External Network
Front-end
Cluster Computer Architecture
SSI
• A single system image is the illusion, created
by software or hardware, that presents a
collection of resources as one, more powerful
resource.
• SSI makes the cluster appear like a single
machine to the user, to applications, and to the
network
Network
• Network is the most critical part of a cluster
• Its capabilities and performance directly

influences the applicability of the whole
system .
Choice of Network
– Ethernet (10Mbps)
– Fast Ethernet (100Mbps)
– Gigabit Ethernet (1Gbps)
– ATM
– Myrinet(1.2Gbps)
Operating System
– Linux
– Microsoft windows
– SUN Solaris
– IBM AIX
– Mach(microkernal based OS)
– Cluster operating systems(Solaris MC,SCO
Unixware)
Parallelism
High performance in cluster is mainly achieved
through parallelism.
Parallelism operates at two levels:
 Hardware parallelism
 Software parallelism
Hardware parallelism
 Hardware parallelism deals with the CPU of an individual
system and how we can squeeze performance out of sub-
components of the CPU that can speed up our code.
 fine grained and coarse grained parallelism.
Software parallelism
 Software parallelism is the ability to find well defined areas
in a problem we want to solve that can be broken down into self-
contained parts .
. These parts are the program elements that can be
distributed and give us the speedup that we want to get out of a
high performance computing systems.
Cluster Components
Beowulf Cluster
Beowulf architecture is a multi-computer
architecture that is used for parallel computation
applications. Primarily, a Beowulf cluster consists
of a server computer that controls the functioning
of many client nodes that are connected together
with Ethernet or any other network comprising of a
network of switches or hubs.
Cluster Applications
Google Search Engine.
Petroleum Reservoir Simulation.
Protein Explorer.
Earthquake Simulation.
Image Rendering.
Banking
E-commerce
Data mining
Issues to be considered
• Cluster networking
• Speed selection
• Timing
An example of Cluster
FUTURE TRENDS
Grid computing
The Grid is a large system of computing resources
that performs tasks and provides to users
a single point of access, commonly based on the
World Wide Web interface, to these distributed
resources.
Users can submit thousands of jobs at a time
without being concerned about where they run.
The grid may scale from single systems to
supercomputers that utilize thousands of
processors.
Conclusion
Clusters are promising:
Offer incremental growth

New trends in hardware and software
technologies are likely to make clusters more
promising and fill SSI gap.
Clusters based supercomputers (Linux based
clusters) can be seen everywhere!
Reference
• www.ieeetfcc.org
• www.buyya.com
• www.clustercomp.org
• www.sgi.com
• http://en.wikipedia.org/wiki/Computer_cluster.ht
m

Presented By: Veena.K.P Mca S5 Roll No:28

Uploaded by

Copyright:

Available Formats

Presented By: Veena.K.P Mca S5 Roll No:28

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Presented By: Veena.K.P Mca S5 Roll No:28

Uploaded by

Copyright:

Available Formats

Presented by:

Cluster Computing is the mechanism of

A cluster is a collection of connected,

• Many standalone computers (nodes).

• High cost of ‘traditional’ High Performance

 provides uninterrupted availability of data

 distributes incoming requests for resources

 multiple processors share a single memory

They hide most of the cluster behind the

compute compute compute compute File

All nodes can be seen from outside,and

High Speed Network

compute compute compute compute File

• Network is the most critical part of a cluster

• Its capabilities and performance directly

Parallelism operates at two levels:

Offer incremental growth

You might also like