[Fwd: Benchmarking Beowulf]

[Mark Dalton]

> 
> Hi,
> 
> Any one knows what is the peak floating point performance for Pentinum II?
> (in Mflop/s). Please help.
> 
You may want to define this more.. (either way I am not sure of this, 
peak is fairly irrelevant compared to actual performance).
>
> Also please advise me how to benchmarking Beowulf? What are the common
> benchmark used ?
> 

I think that someone covered this very well just a few e-mails ago.
It depends...
Here are some benchmarks that were done in the ACL (Advanced Computing Lab)
at LANL (Los Alamos National Labs) cluster:
    http://www.acl.lanl.gov/cluster/benchmarks.html
	* Note the bottom of the page has some graphs comparing performance
	  versus "some companies" machine.

	- Network benchmarks
	      * ping (simple but it gives a idea)
	      * netpipe
	      * mpbench
	- Compute (processor/cache) benchmarks
	      * NAS
	- Memory bandwidth
	      * Streams
	- I/O benchmarks (all-to-one, or parallel, etc.)
	      * ??

Of course you benchmark should be ones that reflect your needs, and
_always_ test your code if posible, because in the end that is what
really matters.  (or simulate the type of use, you want/need/plan to do).

Mark
-- 
Mark Dalton       CH3-S-CH2 H                      H      O       H
Silicon Graphics, Inc.  |   |                      |       \      |
Eagan, MN 55121         CH2-C-COO    //\ ---C--CH2-C-COO    C-CH2-C-COO
mwd@sgi.com                 |       |  ||   ||     |       //     |
                            NH3      \\/ \ / CH    NH3    O       NH3
                                          NH
My home page: http://www.cbc.umn.edu/~mwd/mwd.html
Cell Biology: http://www.cbc.umn.edu/~mwd/cell.html