[mpiwg-persistence] Meeting today

[Richard Graham]

I am flying today, so will not be able to make this call.  I will plan on joining the next call.

Rich



Sent via the Samsung Galaxy S9, an AT&T 5G Evolution capable smartphone

-------- Original message --------
From: "Skjellum, Anthony" <Tony-Skjellum at utc.edu>
Date: 8/7/19 03:42 (GMT-08:00)
To: mpiwg-persistence at lists.mpi-forum.org
Cc: "Grant, Ryan Eric (-EXP)" <regrant at sandia.gov>, Richard Graham <richardg at mellanox.com>
Subject: Meeting today

Hi, we will meet today to discuss

0) Remind ourselves about the Info key ticket and what needs to be done for September
(what I owe, and any help available)
1) Open issues for persistent collective communication
2) Partitioned point-to-point proposal/ticket
3) On-going discussion of making initialization functions local for persistent collectives, how that impacts other parts of the standard (Dan/Tony/Rolf thread of discussion in past that's been renewed lately)
4) Rich Graham's idea(s) for possibly changing base addresses in persistent collectives without a new initialization ; INFO key? there are strong implications on optimization from this degree of freedom.  [If Rich can't make the call, we will just skim the surface and
continue this when he can join.]

Anything else?

Thanks,
Tony Skjellum



Anthony Skjellum, PhD

Professor of Computer Science and Chair of Excellence

Director, SimCenter

University of Tennessee at Chattanooga (UTC)

tony-skjellum at utc.edu  [or skjellum at gmail.com]

cell: 205-807-4968

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-persistence/attachments/20190807/b8108c71/attachment.html>