Hi everyone,
       As we plan out the details of a proposal for conference funding for the fall workshops, we need to determine some basic parameters for a budget, after which everything else goes smoothly, as conference proposals need not be more than 2-3 pages.  I'd like to begin an e-mail discussion this afternoon with whomever is interested in helping to determine our budget needs, but to avoid spamming people, can anyone who wishes to join in e-mail me?  Tanja and Roland are automatically included since er are planning on holding it in Atlanta...

Cheers,
Josh

On Tue, Jul 12, 2011 at 10:52 AM, Roland Haas <roland.haas@physics.gatech.edu> wrote:
Meeting minutes:

Present were: Tanja, Roland, Frank, Josh, Bruno, Yosef, Peter, Erik, Eloisa

status of ET paper:
no progress everbody was busy with other things.

fall ET workshop:
* Josh put his draft for a proposal into svn (https://svn.einsteintoolkit.org/documents/Proposal_WS_Fall2011/)

Lonestar issues:
* runs with large number of cores (>=72) hang often
* happens to RIT and GT using the simfactory options list (but at least in the case of GT our own submission script)
* known to happen (Frank knows)
* happens with git version of Cactus
* GT: when hang occurs then the processes hang in Carpet/Recompose (specifically a comm_state routine, I think step) where most processes think they are done with a refinement level but some are not yet done and wait for more data.
* suspect some MPI related problem
* will try differnt MPI versions and all contact the help desk

Yours,
Roland

--
My email is as private as my paper mail. I therefore support encrypting
and signing email messages. Get my PGP key from http://keys.gnupg.net.


_______________________________________________
Users mailing list
Users@einsteintoolkit.org
http://lists.einsteintoolkit.org/mailman/listinfo/users