Set of loosely connected Computers that work together so
that in many respects they can be viewed as a single
system
Enable scientists to concentrate on science rather than on
computing
docsity.com
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Guidelines and tips
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Clusters-Parallel Processing-Slides, Slides of Parallel Computing and Programming
This project report was submitted to Prof. Sanabhi Sarin at Ankit Institute of Technology and Science for Parallel Processing. It includes: Concentrate, Oscar, Rocks, Source, Cluster, Application, Resources, , Installation, Administration, Performance
Typology: Slides
2011/2012
1 / 47
This page cannot be seen from the preview
Don't miss anything!
Related documents
Partial preview of the text
Download Clusters-Parallel Processing-Slides and more Slides Parallel Computing and Programming in PDF only on Docsity!
Set of loosely connected Computers that work together so that in many respects they can be viewed as a single system Enable scientists to concentrate on science rather than on computing
Oscar
OpenMosix
Rocks
The installation of OSCAR must be done on
an already working Linux machine
Its based on SIS and DHCP server
A single-image cluster has multiple copies of
a single operating system kernel
Users can start from any node in the cluster
OpenMosix starts a process on one machine
and actually runs on another machine in the
cluster
Each process has its own Unique Home Node
(UHN) where it gets created
Process are made of
two pieces
Deputy Remote
Inherit granularity attached to process
migration
Lack of scheduling control
Kernel dependent
Shared memory issues
There are issues with Multiple Threads not
gaining performance
Useful to abstract load balancing
Bad for programmatic load balancing, since
the user has no control over how load
balancing is achieved
Make Clusters Easy
Build a cluster without assuming CS knowledge
Results in a very robust system that is insulated from human mistakes
No special “Rocksified” package structure. Can install
any RPM.
Easiest way to build a Cluster Free More supported architectures ▪ Pentium, Athlon, Opteron, Nocona, Itanium Widely Used ▪ 1100 registered clusters, 700 member support list ▪ HPCwire Readers Choice Awards 2004 More configured HPC software: 15 optional extensions (rolls) and counting
Quick to install It should not take a month (or even more than a day) to install a thousand node cluster Nodes are 100% configured No “after the fact” tweaking If a node is out of configuration, just reinstall Don’t spend time on configuration management of nodes Just reinstall Scalability Nodes install from kickstart files generated from a database Several clusters registered with over 500 nodes
Big projects use Rocks
BIRN (20 clusters)
GEON (20 clusters)
NBCR (6 clusters)
Avalanche installer removes pressure from any single installation server Introduced in Rocks 4. Torrent based Nodes share packages during installation