[gmx-users] Parallel Gromacs Benchmarking with Opteron Dual-Core & Gigabit Ethernet

[David van der Spoel]

Erik Lindahl wrote:
> Hi,
> 
>> I read at Gmx site that the DPPC system
>> composed of 121,856 atoms. I saw the gmx topology files, it
>> seems that Gmx makes data decomposition on input data to run in parallel
>> (in our simulation case using "-np 12" for execution
>> on 3 nodes, the data space for every process is about 10156 atoms).
>> I think that the DPPC system's size is not so big enough that someone
>> can sense the scalability of parallel execution in the existence of
>> Gigabit Eth. I mean to see the Cluster scalability in our configuration,
>> we should setup a bigger simulation. Pls correct me, if I'm in mistake.
> 
> Well, you can always try different systems (genconf + edit topology), 
> but the fact that you see low user CPU usage likely means the nodes are 
> busy waiting for the communication (which probably counts as 
> kernel/system usage).
> 
> Remember - compared to the benchmark numbers at www.gromacs.org, your 
> bandwidth is 1/4 and the latency 4 times higher, since you have four 
> cores sharing a single network connection.
> 
> Gromacs 4 should scale better on any hardware (significantly better with 
> PME), but you'll probably never see great scaling with only 4-way shared 
> gigabit ethernet.  It's available in the head branch of CVS for expert 
> users/voluntary guinea-pigs, but entirely unsupported until we release it.
> 
in addition with gromacs 3.3 you want to use the -shuffle option for grompp.


-- 
David van der Spoel, Ph.D.
Molec. Biophys. group, 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