[Beowulf] Strange Opteron 2350 performance: Gaussian-03
lindahl at pbm.com
Sat Jun 28 15:54:12 PDT 2008
On Sun, Jun 29, 2008 at 02:30:54AM +0400, Mikhail Kuzminsky wrote:
> (BTW, there is one bad thing for stream on this server - the
> corresponding data are absent in McCalpin's table: the throughput is
> scaled good from 1 to 2 OpenMP threads, and gives good result for 8
> threads, but the throughput for 4 threads is about the same as for 2
> threads. The reason is, IMHO, that for 8 threads RAM is allocated by
> kernel in both nodes, but for 4 threads the RAM allocated is placed in
> one node, and 4 threads have bad competition for memory access).
Er, this is not a general result, but is a function of your OpenMP
implementation. We just discussed it a couple of days ago, right here.
More information about the Beowulf