[gmx-developers] interesting read: exascale w/o threads
David van der Spoel
spoel at xray.bmc.uu.se
Thu Oct 8 15:08:18 CEST 2015
On 08/10/15 13:36, Berk Hess wrote:
> Interesting read. The essential message is:
> "Thus, we believe that a discussion on threads versus processes boils
> down to “shared everything by default” versus “shared nothing by default”.
> We came to the same conclusion in a discussion some time ago. So the
> choice doesn't affect performance, but it strongly affects the code.
> For Gromacs I think it's still convenient to have processes+threads,
> since we have many data structures with many small arrays that change at
> domain decomposition time that are needed by all threads in a domain.
> Sharing all these through MPI3 is tedious.
Also I think it can be advantageous to have fewer DD cells in case of
very inhomogoneous systems (we are working with clusters in the gas phase).
> On 10/08/2015 01:04 PM, Szilárd Páll wrote:
>> "We believe that portions of the HPC community have adopted the point
>> of view that somehow threads are “necessary” in order to utilize such
>> [manycore/SMP] systems, (1) without fully understanding the
>> alternatives, including MPI 3 functionality, (2) underestimating the
>> difficulty of utilizing threads efficiently, and (3) without
>> appreciating the similarities of threads and processes. This short
>> paper, due to space constraints, focuses exclusively on issue (3)
>> since we feel it has gotten virtually no attention."
David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone: +46184714205.
spoel at xray.bmc.uu.se http://folding.bmc.uu.se
More information about the gromacs.org_gmx-developers