On Tue, Jul 14, 2020, Jonathan Lefman wrote:
>
>I think the group in Italy wants to run Amber20 on a Power9 GPU
>system. Do you think we should introduce them to you so you can help them
>meet the person from IBM who made the patch for Amber20? I’m not sure if
>Ruzhu ever connected with you on maintaining their Power9 patch in the
>Amber repo. What do you recommend?
>
>Jon
>
>From: Filippo Spiga <fspiga.nvidia.com>
>
>CINECA in Italy deployed a bit Marconi 100. They have users looking at
>building AMBER 20 with XLC and Spectrum MPI. I am aware of a patch that
>makes this possible, who is the best person within the AMBER development
>group to provide such patch to AMBER licensee?
>
>—
>FILIPPO SPIGA
>EMEA HPC Developer Relations
>
I'm cc-ing your message to the Amber developer's group. I think(?)
people have managed to get all of Amber working on a power9/GPU system
called "Summit" at Oak Ridge, but I don't know anything about the status
of that. I don't recall any patches being posted to our development
platform on this subject. Ruzhu Chen has an account at
gitlab.ambermd.org, but I don't think has made merge requests.
I could be mis-remembering here. Developers: anyone have info to share
about power9?
I'm just an intermediary here, and have no real answers to share.
...thx...dac
--
====================================================================
David A. Case | david.case.rutgers.edu
Dept. of Chemistry & Chemical Biology |
Rutgers University | office: +1-848-445-5885
174 Frelinghuysen Road, Rm. 208b | cell: +1-609-751-8668
Piscataway, NJ 08854 USA | web: casegroup.rutgers.edu
====================================================================
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Tue Jul 14 2020 - 08:30:03 PDT
