[gmx-users] Benchmarks vs memory bandwidth (Mike Sullivan)
schraven at mpip-mainz.mpg.de
Tue May 18 21:52:30 CEST 2004
Maybe it's a good idea to do the same test with the big dppc system, as it
eats a lot more memory, this might increase the difference between the
p.s. apologies to the list for the bad title of my previous post, when
replying on it, please be so kind to change it into something better.
Pim Schravendijk - PhD Student
Max Planck Institute for Polymer Research
> Message: 9
> Date: Tue, 18 May 2004 15:42:43 -0400
> From: Mike Sullivan <mike.sullivan at alltec.com>
> Subject: [gmx-users] Benchmarks vs memory bandwidth
> To: gmx-users at gromacs.org
> Message-ID: <40AA6733.5010206 at alltec.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> I ran a few tests to determine the importance of memory bandwidth.
> The results are rather
> interesting. The test case is a 50ps simulation that was given to me by
> a student. The machine
> is a Dual Opteron 248 ( 2Ghz ) based on a Tyan S2885 using 1GB DDR 400
> ECC Reg modules.
> This was to determine if the Tyan 2875 was a viable gromacs platform.
> It has only
> 4 dimm sockets that are attached to one of the two cpus. The second cpu
> must access the
> memory pool via the first. You get only two memory channels as opposed
> to four on the
> 288x series.
> Suprisingly even with a single memory channel the performance is similar
> to the
> full bandwidth run.
> 4 memory channels ( 2 dimm Per cpu ) 23:06
> 2 memory channels (2 on a single cpu ) 23:24
> 1 memory chanel ( 1 dimm on single cpu ) 23:37
> FYI for the last case
> NODE (s) Real (s) (%)
> Time: 1417.000 1417.000 100.0
> (Mnbf/s) (GFlops) (ps/NODE hour) (NODE hour/ns)
> Performance: 103.784 3.102 127.029 7.872
> This was using the 3.2.1 rpms and run on Suse Enterprise 8.0 for AMD64
> Mike Sullivan V.P. Performance Computing
> @lliance Technologies, Voice: (416) 385-3255 x 228,
> 18 Wynford Dr, Suite 407 Fax: (416) 385-1774
> Toronto, ON, Canada, M3C-3S2 Toll Free:1-877-216-3199
More information about the gromacs.org_gmx-users