[gmx-developers] threads are now ON by default

Sander Pronk pronk at cbr.su.se
Fri Feb 12 18:20:27 CET 2010


Those rumors are in fact true; according to Berk there's a significant speedup when using 8 threads on a Core i7 - I just wrote Ncores because I didn't know of a good word for 'hardware point of execution on a cpu'. 

As an aside, to test performance on a Core i7 without needing to turn hyperthreading on or off in BIOS, it's best to use taskset (in Linux); use

taskset 0x0f mdrun -nt 4 

to run 4 threads on 4 different physical cores, and just use

mdrun -nt 8 
taskset 0xff mdrun -nt 8

to run 8 threads on all the HT-enabled cores (or 'hardware points of execution').


On 12 Feb 2010, at 18:14 , David van der Spoel wrote:

> On 2/12/10 5:32 PM, Sander Pronk wrote:
>> Now that the last issues have been resolved with the threading code, thread-based parallelization has been turned on by default. To disable all the threading code, use
>> --disable-threads.
>> in configure, or turn the option GMX_THREADS off with ccmake.
>> Running mdrun with just one thread (the default) is almost exactly the same as running it without threading code: the only thing that's different is that the few remaining global variables are protected by mutexes.
>> Performance-wise mdrun runs very slightly faster with threads than with OpenMPI when Nthreads<=Ncores (and there is no other processes on the computer). When Nthreads>Ncores (or other processes are running), the thread code is much faster than OpenMPI, but the total runtime is still smaller than when Nthreads==Ncores.
>> If there's any problems in getting things running, or with performance, I'd very much like to hear about it.
>> Sander
> Great work Sander!
> What about the rumours that enabling hyperthreading gives you superscaling beyond Ncores?
> -- 
> David van der Spoel, Ph.D., Professor of Biology
> Dept. of Cell & Molec. Biol., Uppsala University.
> Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205. Fax: +4618511755.
> spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
> -- 
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-developers
> Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-developers-request at gromacs.org.

More information about the gromacs.org_gmx-developers mailing list