[yt-dev] rockstar inline with enzo

[Matthew Turk]

Hi Stephen,

Yes -- my recollection is actually that when running rockstar inline,
we have three models.  One is a sidecar set of processes which are
resident either in the same MPI_COMM_WORLD but a different Enzo / yt
top communicator.  Another would be a second set of processes launched
on a different set of jobs (one advantage of the non-MPI communication
used in Rockstar.)  I think the best one though would be to have
Rockstar's readers and writers operate on the same nodes, in the same
MPI process space.  This will require giving up the GIL on the writer
jobs and allowing them to wait, then reclaiming it once the readers
have done their business.

If you watch Peter's talk at HIPACC, he goes into some discussion of
how that final mode operates, andi t's actually the one he recommends.
 Or so I recall.

-Matt

On Wed, Nov 14, 2012 at 5:13 PM, Stephen Skory <s at skory.us> wrote:
> Hi all (mainly Matt),
>
> I know we have several threads open on Rockstar and halo finding
> already, but I'm wondering how Rockstar is going to work inline with
> Enzo? It seems to me that every MPI thread will need to be a "reader,"
> but that doesn't leave room for the "writers" or the "server". Matt,
> you mentioned to me that you had thought about this already. I'm
> wondering if you could share your ideas? I'd love to help make it
> work!
>
> Thanks!
>
> --
> Stephen Skory
> s at skory.us
> http://stephenskory.com/
> 510.621.3687 (google voice)
> _______________________________________________
> yt-dev mailing list
> yt-dev at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org