charmm scalability on 2.4 kernels
David van der Spoel
spoel at xray.bmc.uu.se
Tue Jan 8 10:51:27 PST 2002
On Tue, 8 Jan 2002, Josip Loncaric wrote:
>Actually, Linux kernel 2.4 incorporates an improved yet very different
>TCP stack where my TCP fixes do not help much in my usual test case
>(point-to-point streaming of small messages). However, Steve's
>scalability problems with the stock 2.4 TCP are very interesting, since
>they involve many machines, i.e. a completely different communication
>pattern from my usual point-to-point tests.
>Old 2.2+fix combination was pretty efficient at aggregating small
>messages into larger TCP packets before sending, in fact better than
>stock 2.4. Packet aggregation is something that depends on delicate
>timing of congestion control events on both sender and receiver; this is
>very sensitive to the application's communication pattern. Perhaps a
>2.4 TCP fix would need to be developed after all...
Actually for molecular dynamics I think the problem is mainly latency.
Scali/Myrinet is a big win for large number of processors (see
www.gromacs.org/benchmarks/scaling.php for my own benchmarks).
However for me it also helped quite a bit to increase the TCP and Shared
memory short message size for LAM (I use 512 kb). Apparently short and
long messages are treated differently.
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
More information about the Beowulf