0

If you guys have never heard of the idea behind Condor Cluster: http://www.cs.wisc.edu/condor/ It's developed by the University of Wisconsin but used by many other universities since its birth.

I was just wondering, in general though, how Cluster Computing actually finishes its jobs. Does it take the job that you submitted, and split it into pieces and allocate the different pieces to multiple workstations scattered across the cluster? Or does it use just one of those workstations and run your whole task solely as a foreground job?

Any insight on this whole cluster/grid computing concept would be very helpful. Thanks

Dark Templar
  • 97
  • 4

2 Answers2

2

This is very much dependent on the implementation at hand (both of the cluster and the actual computer job), which in turn is dependent on the type of problems it tries to solve.

There are computing problems that are impossible to compute in parallel systems, while others have demand for extremely fast IPC or are independent of each other and scale quasi linear with the number of nodes.

Sven
  • 97,248
  • 13
  • 177
  • 225
  • Thank you, SvenW. Perhaps do you know what some of these impossible to compute in parallel-environments problems would be? – Dark Templar Oct 09 '11 at 18:26
  • I can't think of anything specific right now, but every problem where each step depends on the previous one fits into this category. – Sven Oct 09 '11 at 18:46
0

After researching stuff about clusters I came across Amdahl's law:

used to find the maximum expected improvement to an overall system when only part of the system is improved [part that can be done concurrently]. It is often used in parallel computing to predict the theoretical maximum speedup using multiple processors.

Michael Hampton
  • 237,123
  • 42
  • 477
  • 940
Hax
  • 139
  • 6