September 2018 - mpiwg-sessions - mailman3.cels.anl.gov

MPICH/hydra happier with Dan's test cases (kind of)
by Pritchard Jr., Howard 16 Sep '18

16 Sep '18

HI Folks, MPICH/hydra is happy with libCFG_noMCW, even with all the sleeps removed. I ran up to 36 ranks with hydra/mpich and didn’t see a problem. Its not so happy with the other tests, I think they are buggy. Here’s what I get for libCFG_noMCW_multiport hpp@sn-fey1:/usr/projects/hpctools/hpp/mpi_sessions_code_sandbox>mpiexec -n 8 ./libCFG_noMCW_multiport process 1 (MPI_COMM_WORLD) now calling barrier on MPI_COMM_WORLD process 3 (MPI_COMM_WORLD) now calling barrier on MPI_COMM_WORLD process 4 (MPI_COMM_WORLD) calling from group thingy rank 4 non-trivial use-case: target group size 4, localGroup size 1 rank 4 opening port process 5 (MPI_COMM_WORLD) now calling barrier on MPI_COMM_WORLD process 6 (MPI_COMM_WORLD) calling from group thingy rank 6 non-trivial use-case: target group size 4, localGroup size 1 rank 6 opening port process 7 (MPI_COMM_WORLD) now calling barrier on MPI_COMM_WORLD process 0 (MPI_COMM_WORLD) calling from group thingy rank 0 non-trivial use-case: target group size 4, localGroup size 1 process 2 (MPI_COMM_WORLD) calling from group thingy rank 2 non-trivial use-case: target group size 4, localGroup size 1 rank 2 opening port rank 2 opened port tag#0$description#sn-fey1.lanl.gov$port#55566$ifname#128.165.227.181$ rank 2 publishing port tag#0$description#sn-fey1.lanl.gov$port#55566$ifname#128.165.227.181$ using name foobar10 round 1 rank 4 opened port tag#0$description#sn-fey1.lanl.gov$port#45508$ifname#128.165.227.181$ rank 4 publishing port tag#0$description#sn-fey1.lanl.gov$port#45508$ifname#128.165.227.181$ using name foobar10 round 2 rank 6 opened port tag#0$description#sn-fey1.lanl.gov$port#52388$ifname#128.165.227.181$ rank 6 publishing port tag#0$description#sn-fey1.lanl.gov$port#52388$ifname#128.165.227.181$ using name foobar10 round 3 rank 2 published port tag#0$description#sn-fey1.lanl.gov$port#55566$ifname#128.165.227.181$ using name foobar10 round 1 rank 2 accepting on port tag#0$description#sn-fey1.lanl.gov$port#55566$ifname#128.165.227.181$ (localSize 1) Fatal error in PMPI_Publish_name: Invalid service name (see MPI_Publish_name), error stack: PMPI_Publish_name(134): MPI_Publish_name(service="foobar10 round 2", MPI_INFO_NULL, port="tag#0$description#sn-fey1.lanl.gov$port#45508$ifname#128.165.227.181$") failed MPID_NS_Publish(67)...: Lookup failed for service name foobar10 round 2 Fatal error in PMPI_Publish_name: Invalid service name (see MPI_Publish_name), error stack: PMPI_Publish_name(134): MPI_Publish_name(service="foobar10 round 3", MPI_INFO_NULL, port="tag#0$description#sn-fey1.lanl.gov$port#52388$ifname#128.165.227.181$") failed MPID_NS_Publish(67)...: Lookup failed for service name foobar10 round 3 If I have a chance I”ll play with this test and see if I can get it to work. hmm.. maybe hydra doesn’t like the whitespace publish names. Howard -- Howard Pritchard B Schedule HPC-ENV Office 9, 2nd floor Research Park TA-03, Building 4200, Room 203 Los Alamos National Laboratory

1 0

Re: [mpiwg-sessions] joining a little late today
by Tony Skjellum 10 Sep '18

10 Sep '18

I’m just arriving in UK for cluster meeting — will not be able to Join — can there be minutes ? Thx Tony Anthony Skjellum, PhD 205-807-4968 > On Sep 10, 2018, at 4:52 PM, Pritchard Jr., Howard via mpiwg-sessions <mpiwg-sessions(a)lists.mpi-forum.org> wrote: > > HI Folks, > > I’ll be about 10 minutes late today to WG webex. > > Howard > > -- > Howard Pritchard > B Schedule > HPC-ENV > Office 9, 2nd floor Research Park > TA-03, Building 4200, Room 203 > Los Alamos National Laboratory > > _______________________________________________ > mpiwg-sessions mailing list > mpiwg-sessions(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpiwg-sessions

1 0

Re: [mpiwg-sessions] joining a little late today
by HOLMES Daniel 10 Sep '18

10 Sep '18

Hi Howard, I’m back in the land of the internet-equipped, so I’ll join the call in 1 hour. All - connection details can be found here: http://mpi-forum.epcc.ed.ac.uk/mpiwg-sessions.php Cheers, Dan. — Dr Daniel Holmes PhD Applications Consultant in HPC Research d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 10 Sep 2018, at 16:52, Pritchard Jr., Howard via mpiwg-sessions <mpiwg-sessions(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: HI Folks, I’ll be about 10 minutes late today to WG webex. Howard -- Howard Pritchard B Schedule HPC-ENV Office 9, 2nd floor Research Park TA-03, Building 4200, Room 203 Los Alamos National Laboratory _______________________________________________ mpiwg-sessions mailing list mpiwg-sessions(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-sessions The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 0

joining a little late today
by Pritchard Jr., Howard 10 Sep '18

10 Sep '18

HI Folks, I’ll be about 10 minutes late today to WG webex. Howard -- Howard Pritchard B Schedule HPC-ENV Office 9, 2nd floor Research Park TA-03, Building 4200, Room 203 Los Alamos National Laboratory

1 0

4 PRs open now for sessions
by Pritchard Jr., Howard 03 Sep '18

03 Sep '18

Hi Dan, Four PRs open now for Sessions: * changes to chapters 6 and 8 * switch to “pset” notation in chapter 10, introduce MPI info like query for process set names. * add an example to chapter 10 (haven’t verified it compiles yet) * update chapter 6 with MPI_GROUP _FROM _SESSION _PSET (I didn’t like the name in the slide deck) and some verbiage changes at the beginning of group constructor section These changes don’t include corresponding change log entries. I was hoping you could do that as part of a grand merge of all these separate PRs into the big one for the Forum. I am going through the standard in general looking for places that need to be modified in light of the Peer to Peer model. I’d say for purposes of expediting, you go ahead and make changes to the current 4 PRs as you see fit as your merge them into the final PR. Note Tuesday morning between 5 and 7 PM your time I’ll be behind the fence and won’t have cell contact, may have email but don’t count on it. Wednesday morning I’ll again be behind the fence between 5 and 7 PM your time, so same story as far as reachability. Howard -- Howard Pritchard B Schedule HPC-ENV Office 9, 2nd floor Research Park TA-03, Building 4200, Room 203 Los Alamos National Laboratory

1 0

meet today
by Pritchard Jr., Howard 03 Sep '18

03 Sep '18

Hi Folks, Despite being a government holiday in UK and US, we plan to meet to go over PRs at https://github.com/mpiwg-sessions/mpi-standard/pulls We need to have these merged into one, plus updates to change log, etc. in preparation for the next forum. Howard -- Howard Pritchard B Schedule HPC-ENV Office 9, 2nd floor Research Park TA-03, Building 4200, Room 203 Los Alamos National Laboratory

1 0

Re: [mpiwg-sessions] PMIx Group status update
by Ralph H Castain 02 Sep '18

02 Sep '18

Hi folks I have completed the updates to both the web site description and the RFC: https://pmix.org/pmix-standard/pmix-groups/ https://github.com/pmix/RFCs/pull/33 The RFC references the prototype code. There is still a bug or two in the PMIx_Group_invite implementation, and so I haven’t formally submitted the PR yet. Please take a look and let me know what you think. I’ll be out on vacation for a few days, but will pick up the implementation again when I return. Ralph > On Aug 29, 2018, at 2:08 PM, Pritchard Jr., Howard <howardp(a)lanl.gov> wrote: > > Thanks very much Ralph! > > -- > Howard Pritchard > > B Schedule > HPC-ENV > Office 9, 2nd floor Research Park > TA-03, Building 4200, Room 203 > > Los Alamos National Laboratory > > > > > > On 8/29/18, 12:29 PM, "mpiwg-sessions on behalf of Ralph H Castain via > mpiwg-sessions" <mpiwg-sessions-bounces(a)lists.mpi-forum.org on behalf of > mpiwg-sessions(a)lists.mpi-forum.org> wrote: > >> Hi folks >> >> I promised on the last call to keep people apprised of my progress. I >> have implemented the collective versions of the new APIs >> (PMIx_Group_construct/destruct) and have updated the PMIx_Fence operation >> to look for group names, substituting the group members into the array of >> participants. This all appears to be working correctly. >> >> The PR for it is here: https://github.com/pmix/pmix/pull/826 >> or the branch itself: >> https://github.com/rhc54/pmix-master/tree/topic/group >> >> You can see an example use of the code here: >> https://github.com/rhc54/pmix-master/blob/topic/group/examples/group.c >> >> I’m now working on the async APIs (PMIx_Group_invite/join) - I have them >> mostly implemented, but need to debug. >> Ralph >> >> _______________________________________________ >> mpiwg-sessions mailing list >> mpiwg-sessions(a)lists.mpi-forum.org >> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-sessions >

1 0