[mpich-discuss] Existence of something like MPI_Type_compare()?

[Jed Brown]

On Tue, Dec 13, 2011 at 11:23, Dave Goodell <goodell at mcs.anl.gov> wrote:

> Also, if you only need to check for datatype compatibility and not all of
> the other issues listed in our paper, you probably would benefit from the
> hashing scheme developed by Langou et al.:
> http://www.springerlink.com/content/6xdbmmva33cwvj33/
>
> I'm not sure how to obtain a copy of that code though at this point.  The
> listed web page for downloading it is offline and archive.org does not
> seem to have a copy.
>

Hmm, this looks neat, but since my check for equivalence is necessary for
correctness (rather than just error detection), it's important to not have
collisions. A hash could be used to quickly identify negatives, but I would
still need a collision-free check eventually.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/mpich-discuss/attachments/20111213/c5efb338/attachment.htm>