[gmx-users] Running with > 5 NODES

[Chng Choon Peng]

Hi David,

   I just ran fftw_mpi_test with up to 16 CPUs (on our 8 dual-CPU nodes) 
and it worked fine. Tested both 128 and 256 FFT sizes. The problem 
probably does not lie there.

regards,
choon peng

David van der Spoel wrote:

>On Fri, 4 Jan 2002, Chng Choon Peng wrote:
>
>Yes but how about FFTW? Isn't there a test program that comes with FFTW?
>Either way, scaling of the PME code is quite poor over more than a few
>nodes (2-3)
>
>Groeten, David.
>________________________________________________________________________
>Dr. David van der Spoel, 	Biomedical center, Dept. of Biochemistry
>Husargatan 3, Box 576,  	75123 Uppsala, Sweden
>phone:	46 18 471 4205		fax: 46 18 511 755
>spoel at xray.bmc.uu.se	spoel at gromacs.org   http://zorn.bmc.uu.se/~spoel
>++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
-- 
Chng Choon Peng
Scientific Programmer
BioInformatics Institute (BII)
30 Medical Drive, Level 1 IMCB Building, Singapore 117609
Tel (O): 8746173