Table of contents
    No headers
    To present an overview of the GROMACS molecular dynamics performance and compare the simulation speed attainable on some of the most common hardware platforms we have constructed a benchmark of a few typical systems. The benchmarks all represent "real-life" examples, i.e. they have been taken from ongoing research projects either in our labs or from published articles.


    You can download our entire set of benchmark systems (it hasn't changed for version 3.1):



    The benchmarks are divided in two parts; the first concerns performance on single node machines (i.e. no parallelization outside the machine, but possibly using several CPUs), while the second one shows the scaling capabilities for a large membrane system when running in parallel over an external network.
    Performance is quoted as picoseconds of simulation performed per day for all benchmarks, and the parallel scaling over N processors is defined as


    S = p_N / ( N * p_1 )


    where p_x is the performance on x processors.
    Page last modified 01:04, 13 Oct 2009 by JLemkul?