[gmx-users] Is Total Test time for regression tests relevant to real performance?

Mark Abraham mark.j.abraham at gmail.com
Thu Apr 7 13:41:06 CEST 2016


Lots of factors contribute to performance, but in no case is total time for
the regression tests relevant to real-world performance.


On Wed, Apr 6, 2016 at 11:22 PM Bin Liu <fdusuperstring at gmail.com> wrote:

> Hi All,
> Today I compiled GROMACS 5.1.2 on two machines. One has i7-3770k, the other
> one has i7-4770k. The odd thing I saw was it took 264.77 sec to complete
> the regression tests on the 4770k machine while it took 150.05 sec on the
> 3770k machine. The specs of the 3770k machine are lower in all aspects. The
> gcc for 3770k is 4.6.4, while 4770k is 4.7.3.
> In theory, the total test time for these two machines should be comparable.
> I don't understand what caused such a huge difference. And I am more
> concerned about if the total test time is relevant to real world
> performance.
> Thank you for any clues.
> Regards,
> Bin
> --
> Gromacs Users mailing list
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.

More information about the gromacs.org_gmx-users mailing list