Re: benchmarking of NAMD: sustained performance (TF)

From: Philip Peartree (P.Peartree_at_postgrad.manchester.ac.uk)
Date: Thu May 29 2008 - 04:35:50 CDT

Not wishing to question your methods, but isn't seconds/MD step more
relevant than Flops? Anyhow there are a number of benchmark systems
available here:

http://www.ks.uiuc.edu/Research/namd/utilities/

under example simulations. As far as I'm concerned, when benchmarking,
it is more important to compare representative workloads across the
multiple systems, see:

http://www.ks.uiuc.edu/Research/namd/performance.html

for an interesting graph which may help you.

Philip Peartree
University of Manchester

Quoting satya work <satya.work_at_gmail.com>:

> Dear NAMD users
>
> We are in the process of buying a 5-7 TF cluster and NAMD is one of the
> codes that is routinely used in our group. Are there any benchmarks
> available for NAMD performance in terms of flops for various
> machines/vendors? I have looked in the mailing list and also on the net,
> but could not get the right information. Most of the benchmarking results I
> found are in terms of seconds/MD step. I am interested in sustained
> performance of NAMD in terms of Teraflops (on various architectures) and I
> am wondering if such information has been compiled.
>
> -Thanks
> Satya
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:49:30 CST