[gmx-users] benchmarking results

[David van der Spoel]

vipin sachdeva wrote:
> 
> Hi,
>    I was going over the benchmarking page, for single-node results on 
> different platforms. I don't understand what are the numbers mentioned 
> for each of the benchmarks ? Are they ps/day metric ? Also, what are the 
> compilation flags used in each of the cases ? I would assume "-O3" with 
> SSE/Altivec included for these results ? Were there also system specific 
> options for each of these platforms ? In particular I am interested in 
> ppc970/woodcrest numbers ? How many threads per node are being run for 
> the Woodcrest or the other platforms on the benchmarking page? Is there 
> any benefiton running single/multiple threads on
> multi-core platforms ? Could someone help me with these questions ?
> 
ps/day is correct.
in principle we have hitherto used one thread per physical core, 
ignoring hyperthreading. just try it on your own box.

-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++