CLUSTER COMPUTING

Presented By,
Navaneeth.C.Mouly
1AY05IS037

Under the guidance of,

Prof.Umapathi G.R
 AGENDA
• Abstract.
• Introduction.
• Why Clusters?
• Real Time Example.
• Architecture.
• Cluster Classification.
• Benefits.
• Dark side of cluster computing.
• Applications.
• Challenges.
• Conclusion.
 Abstract
Cluster computing is the technique of
linking two or more computers into a
network (usually through a local area
network) in order to take advantage of
the parallel processing power of those
computers.
 INTRODUCTION
Very often applications need more computing
power than a sequential computer can provide.
One way of overcoming this limitation is to
improve the operating speed of processors and
other components so that they can offer the power
required by computationally intensive applications.
The viable and cost-effective solution is to connect
multiple processors together and co-ordinate their
computational efforts.
 Introduction contd..
Pfister points out, there are 3 ways to
improve performance..
Shared Pool of
Computing
1. Work harder. Resources:
Processors, Memory,
Disks
2. Work smarter
Interconnect
3. Get help
 Why Clusters?
The question may arise why clusters are designed and built when
perfectly good commercial supercomputers are available on the market.
 Clusters are surprisingly powerful .
 They are cheap and easy way to take off-the-shelf components
and combine them into a single supercomputer.
 In some areas of research clusters are actually faster than
commercial supercomputer.
 Clusters also have the distinct advantage that they are simple to
build using components available from hundreds of sources.
 Real Time Example

270 GB RAM

8,700 GB Hard Disk

Pentium 4 Xeon Cluster

Largest Cluster System
IBM BlueGene, 2007

Memory: 73728 GB

OS: CNK/SLES 9

Interconnect: Proprietary

106,496 nodes

478.2 Tera FLOPS on LINPACK

Architecture
 Types Of Clusters

1. High availability or Failover Clusters.

2. Load Balancing Clusters.

3. Parallel/Distributed processing clusters.

Failover Clusters
Load Balancing Cluster
 Benefits Of Clusters

1. Reduced Cost

2. Processing Power

3. Improved Network Technology

4. Scalability

5. Availability
 Dark Side Of Computing
 An eternal struggle in any IT department is in finding a
method to squeeze the maximum processing power out of a
limited budget.
 Today more than ever, enterprises require enormous
processing power in order to manage their desktop
applications, databases and knowledge management .
 Many business processes are extremely heavy users of IT
resources, and yet IT budgets struggle to keep pace with the
ever growing demand for yet more power.
 Challenges
The cluster computing concept also poses three
pressing research challenges:
 A cluster should be a single computing resource
and provide a single system image. This is in
contrast to a distributed system where the nodes
serve only as individual resources.
 The supporting operating system and
communication Mechanism must be efficient
enough to remove the performance Bottlenecks.
 Challenges Cont’d…
 The system’s total computing power should
increase proportionally to the increase in
resources.
 Applications
Few important cluster application are:

1. Google Search Engine.

2. Earthquake Simulation.

3.Image Rendering.
 Google contd…
 Google uses cluster computing to meet the huge
quantity of worldwide search requests that comprise of
a peak of thousands of queries per second.
 A single Google query needs to use at least tens of
billions of processing cycles and access a few hundred
megabytes of data in order to return satisfactory search
result.
 Google Cont’d…

 The first phase of query execution involves index servers consulting an

inverted index that match each query keyword to a matching list of
documents.

 In the second phase, document servers fetch each document from disk to
extract the title and the keyword-in-context portion of the document.

 In addition to the 2 phases, the GWS also activates the spell checker and the
ad server. The spell checker verifies that the spelling of the query keywords is
correct, while the ad server generate advertisements that relate to the query
and may therefore interest the user.
 Conclusion
• Solve parallel processing paradox

• Offer incremental growth and matches with

funding pattern

• New trends in hardware and software

technologies are likely to make clusters more
promising.

• Clusters based supercomputers can be seen

everywhere!
 ?
ANY QUESTIONS?