[AMBER-Developers] first parallel benchmarks (jac) for GPU, S1070 box

From: case <case.biomaps.rutgers.edu>
Date: Tue, 24 Aug 2010 15:45:19 -0400

I have an S1070 box, connected via PCI-16 to a dual-quad Intel "nehalem"
cpu (2.93 GHz). Using the current beta MPI/GPU code and the benchmark
in amber11/benchmarks/jac/bench.jac.cuda (which is not the original jac
benchmark, but what Ross calls the "dhfr production nve" benchmark:

    1 gpu 12.8 ns/day
    2 gpu 19.5 ns/day (76% efficient)
    4 gpu 27.4 ns/day (54% efficient)

For reference, using the cpus on this machine gives:

    2 cores 3.2 ns/day
    4 cores 6.2 ns/day
    8 cores 10.5 ns/day
   16 cores 11.9 ns/day

As usual, please remember that the GPU results are mixed precision, whereas
the CPU results are full double.

I think these are roughly comparable with what Ross and Scott have been
seeing, but they will know better than I do.

I'm hoping that my new box, with 8 C2050 cards, will be delivered soon.


AMBER-Developers mailing list
Received on Tue Aug 24 2010 - 13:00:03 PDT
Custom Search