[Beowulf] Performance characterising a HPC application

Scott Atchley atchley at myri.com
Wed Mar 21 17:37:38 PDT 2007

On Mar 21, 2007, at 5:21 PM, Greg Lindahl wrote:

> On Wed, Mar 21, 2007 at 06:41:07AM -0400, Scott Atchley wrote:
>> I have not benchmarked any applications that need more than 250 MB/s
>> during computation,
> There is a large class of computations which alternate non-overlapped
> compute and communicate cycles. The average over the lifetime of the
> process can be ~ 100 MB/s, but that could easily mean it is sending
> at 1 GByte/s for 10% of the runtime. If you slow down communications
> by a factor of 4, the app will run a lot slower.
> -- greg

I agree that that is a possibility. In the CFD codes that I have  
tested, I have never seen more than 3% difference between  
Myrinet-2000 and Myri-10G, so I assume that the bandwidth is not  
critical in _these_ applications. :-)


More information about the Beowulf mailing list