December 2020 - mpiwg-sessions - mailman3.cels.anl.gov

meet today
by Pritchard Jr., Howard 21 Dec '20

21 Dec '20

Hi Folks, Let’s meet today if possible. Items to discuss * Last minute feedback on dynamic process chapter * Continue with the use cases Martin Schreiber was presenting * Schedule over the holidays Thanks, Howard -- Howard Pritchard HPC-ENV Los Alamos National Laboratory

1 0

Fwd: World/session model
by HOLMES Daniel 18 Dec '20

18 Dec '20

Hi Victor, Thanks for the comments. We’ll discuss them within the WG in the new year. Cheers, Dan. — Dr Daniel Holmes PhD Architect (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. — Begin forwarded message: From: Victor Eijkhout <eijkhout(a)tacc.utexas.edu<mailto:[email protected]>> Subject: Re: World/session model Date: 18 December 2020 at 15:58:20 GMT To: HOLMES Daniel <d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]>> This email was sent to you by someone outside the University. You should only click on links or attachments if you are certain that the email is genuine and the content is safe. My cc to wg-sessions was rejected. I’m resending this in case it didn’t reach you. %%%%%%%%%%%%%%%%% If a program uses both the Sessions model and the World model Thanks. At first I thought the two models were mutually exclusive but I have since found statements that they can be used together. But then I find statements such as 7.2.4 When using the Sessions Model (Section 11.3) for initialization of MPI re- sources, MPI_COMM_WORLD and MPI_COMM_SELF are not valid for use as a communica- tor. ….a little very confusing. To me this sentence says that if I do anything with sessions, the use of MPI_COMM_WORLD is 100 percent ruled out. You mean to say: is ruled out for session-related activities. It certainly confused me. Likewise, in 11.3 there is a very absolute sounding statement “MPI_COMM_WORLD is not valid for use as a communicator.” I’m sure that this sentence comes with an implicit qualification that it applies to using the Session model, but I’m not quite sure how. While I’m nitpicking, 11.2.3 has an ambiguous sentence: “For a library using got the World Model, it needs to know if MPI has been initialized using MPI_INIT or MPI_INIT_THREAD”. The wrong interpretation is “needs to know which of MPI_INIT / MPI_INIT_THREAD” has been used. The correct interpretation (which really took me a minute) is “needs to know if MPI has been initialized using _either_of_ MPI_INIT / MPI_INIT_THREAD” (Oh, and I think “For a library, it needs to know” is not grammatical. Either “A library needs to know” or “For a library, it may be necessary to know”) Victor.

1 0

Re: [mpiwg-sessions] World/session model
by HOLMES Daniel 17 Dec '20

17 Dec '20

Hi Victor, The phrase “not valid for use as a communicator” is carefully chosen and does *not* mean “can’t even exist”. If a program uses both the Sessions model and the World model then that program can use MPI_COMM_WORLD within the context of the World Model but must not use it within the context of the Sessions Model. Thus, there is no conflict between these two statements. The Chapter Committee will take a look at the reference issue that you point out. Thanks for catching that! It had not been raised by the reviewers of that section during the Release Candidate procedure. We will hopefully be able to address it before the Final Release Meeting (Q1 2021). Cheers, Dan. — Dr Daniel Holmes PhD Architect (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 17 Dec 2020, at 15:00, Victor Eijkhout <eijkhout(a)tacc.utexas.edu<mailto:[email protected]>> wrote: This email was sent to you by someone outside the University. You should only click on links or attachments if you are certain that the email is genuine and the content is safe. I’m reading the 4.x standard. I noticed the following: 7.2.4 When using the Sessions Model (Section 11.3) for initialization of MPI re- sources, MPI_COMM_WORLD and MPI_COMM_SELF are not valid for use as a communica- tor. 11.3 MPI objects derived from the Sessions Model shall not be intermixed in a single MPI procedure call with MPI objects derived from the World Model. Are those two at odds? The first statement seems to imply that in the second statement objects from the world model can't even exist. Also: 9.1.2 Environmental Inquiries When using the World Model (Section 11.2), This is the 3rd time the phrase “when using the World Model” appears, but the first time that the reference is given.

1 0

Re: [mpiwg-sessions] Sessions WG - meet today
by HOLMES Daniel 14 Dec '20

14 Dec '20

Hi Howard, Apologies, I’m not going to be able to make it for the Sessions WG call today. Something came up unexpectedly. I approved the MPI_Info_get_string change in the example code. If there’s anything else needed from me, tag me in Github and/or send me email. Cheers, Dan. — Dr Daniel Holmes PhD Architect (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 14 Dec 2020, at 15:21, Pritchard Jr., Howard via mpiwg-sessions <mpiwg-sessions(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: This email was sent to you by someone outside the University. You should only click on links or attachments if you are certain that the email is genuine and the content is safe. Hi All, Let’s meet today to * Discuss followup action items from last week’s Forum, including open PRs and errata tickets * Continue with Martin Schneider’s use cases and diagrams Howard -- Howard Pritchard HPC-ENV 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

1 0

Sessions WG - meet today
by Pritchard Jr., Howard 14 Dec '20

14 Dec '20

Hi All, Let’s meet today to * Discuss followup action items from last week’s Forum, including open PRs and errata tickets * Continue with Martin Schneider’s use cases and diagrams Howard -- Howard Pritchard HPC-ENV Los Alamos National Laboratory

1 0

sessions WG - no meeting today
by Pritchard Jr., Howard 07 Dec '20

07 Dec '20

Hi All, Canceling today’s meeting due to the MPI Forum. Howard -- Howard Pritchard HPC-ENV Los Alamos National Laboratory

1 0