February 2026 - mpi-forum - mailman3.cels.anl.gov

Re: [Mpi-forum] Last Call for Contributions / MPI-Forum F2F / Request to WG Chairs
by Wes Bland 26 Feb '26

26 Feb '26

Hey folks. A bunch of people pinged me this morning about the Eventbrite page closing. Sorry about that. I’ve just extended in-person registration through 5pm EST tomorrow. Please register before that so we can take care of logistics. It looks like the online attendance tickets were also “sold out” and I’ve taken care of that as well. Thanks, Wes > On Feb 19, 2026, at 2:52 PM, Wes Bland via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: > > This Message Is From an External Sender > This message came from outside your organization. > Registration reminder: We’re about a week out from the meeting and there’s a number of people/orgs who I’d expect to have registered at this point, but haven’t. If you’ll be attending please check the website (https://urldefense.us/v3/__https://www.mpi-forum.org/meetings/2026/03/atten… <https://urldefense.us/v3/__https://www.mpi-forum.org/meetings/2026/03/atten…>) to see if your name is where you expect it to be (in-person or remote). > > If you aren’t on that list, please register her ASAP: https://urldefense.us/v3/__https://www.eventbrite.com/e/mpi-forum-march-202… <https://urldefense.us/v3/__https://www.eventbrite.com/e/mpi-forum-march-202…> > > Thanks, > Wes > >> On Feb 16, 2026, at 6:39 AM, Martin Schulz via mpi-forum <mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]>> wrote: >> >> This Message Is From an External Sender >> This message came from outside your organization. >> Hi all, >> >> Our face-to-face meeting in New York is getting near, and today is the two-week deadline for reading and ballot announcements. If you have any, please send them to the list before the end of the day. Also, if there are any other items for a plenary, please let me know. >> >> In general, we are fairly light on official items (kind of expected), and running WG meetings is tricky, as always. On the other hand, we should use the time for some strategic planning towards MPI 6.0: what are the big topics, what is their priority, what should be gating issues, and do we have the right WGs for this? >> >> I will also contact all WG chairs directly, but it would be great if all WGs could prepare a discussion on what should be done in the respective WGs for MPI 6.0 and where we should take it technically. If there are any topics outside the existing WG structure, please let me know as well. >> >> Thanks! >> >> Martin >> >> >> -- >> Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems >> TUM School of Computation, Information and Technology (CIT) >> Dept. of Computer Engineering, TU-Munich, Boltzmannstraße 3, D-85748 Garching >> Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) >> Email: schulzm(a)in.tum.de <mailto:[email protected]> >> >> >> You never fail until you stop trying. (Albert Einstein) >> >> _______________________________________________ >> mpi-forum mailing list >> mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> >> https://urldefense.us/v3/__https://lists.mpi-forum.org/mailman/listinfo/mpi… <https://urldefense.us/v3/__https://lists.mpi-forum.org/mailman/listinfo/mpi…> > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> > https://urldefense.us/v3/__https://lists.mpi-forum.org/mailman/listinfo/mpi…

1 0

Re: [Mpi-forum] Last Call for Contributions / MPI-Forum F2F / Request to WG Chairs
by Wes Bland 19 Feb '26

19 Feb '26

Registration reminder: We’re about a week out from the meeting and there’s a number of people/orgs who I’d expect to have registered at this point, but haven’t. If you’ll be attending please check the website (https://urldefense.us/v3/__https://www.mpi-forum.org/meetings/2026/03/atten… ) to see if your name is where you expect it to be (in-person or remote). If you aren’t on that list, please register her ASAP: https://urldefense.us/v3/__https://www.eventbrite.com/e/mpi-forum-march-202… Thanks, Wes > On Feb 16, 2026, at 6:39 AM, Martin Schulz via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: > > This Message Is From an External Sender > This message came from outside your organization. > Hi all, > > Our face-to-face meeting in New York is getting near, and today is the two-week deadline for reading and ballot announcements. If you have any, please send them to the list before the end of the day. Also, if there are any other items for a plenary, please let me know. > > In general, we are fairly light on official items (kind of expected), and running WG meetings is tricky, as always. On the other hand, we should use the time for some strategic planning towards MPI 6.0: what are the big topics, what is their priority, what should be gating issues, and do we have the right WGs for this? > > I will also contact all WG chairs directly, but it would be great if all WGs could prepare a discussion on what should be done in the respective WGs for MPI 6.0 and where we should take it technically. If there are any topics outside the existing WG structure, please let me know as well. > > Thanks! > > Martin > > > -- > Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems > TUM School of Computation, Information and Technology (CIT) > Dept. of Computer Engineering, TU-Munich, Boltzmannstraße 3, D-85748 Garching > Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) > Email: schulzm(a)in.tum.de <mailto:[email protected]> > > > You never fail until you stop trying. (Albert Einstein) > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org <mailto:[email protected]> > https://urldefense.us/v3/__https://lists.mpi-forum.org/mailman/listinfo/mpi…

1 0

Re: [Mpi-forum] Reading announcement - Clarifying collective synchronisation-semantics
by Tony Skjellum 16 Feb '26

16 Feb '26

Christoph, some thoughts. Good idea to clarify. For v collectives, we have the choice of “must” or “may” synchronize… in the limit of a few non-zero counts, seems clear that May is useful and must is more work. As the MPI application, if the operations may synchronize, the application has to assume it but cannot exploit it. If it is must, then the application can exploit it. In either case, the algorithm can exploit the ability to synchronize. If we allow must to become may for v collectives, we should offer a must version too, since the embedded barrier may be more efficiently done inside the operation. Then the app chooses what is right for them. Neighbor collectives also merit discussion. Example: When implementing v collectives on top of RMA or shmem (persistent mode easiest to use with RMA) we do the cross group sync, even if some counts are zero. If we have further optimizations in persistent mode, we might avoid the whole group… also, likely with neighbor collectives. Regards, Tony Anthony Skjellum, PhD 205-807-4968 On Feb 16, 2026, at 1:32 PM, Christoph Niethammer via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: External Email Warning This email originated from outside the university. Please use caution when opening attachments, clicking links, or responding to requests. This Message Is From an External Sender This message came from outside your organization. Dear all, I would like to present and read PR 1141 "Add text clarifying collective synchronisation-semantics to description" (https://urldefense.us/v3/__https://github.com/mpi-forum/mpi-standard/pull/1…) It implements the feedback from the Oct. 2025 non-voting MPI Forum meeting discussion on the topic raised in issue 971 (https://urldefense.us/v3/__https://github.com/mpi-forum/mpi-issues/issues/9…) Best regards Christoph Niethammer -- Dr.-Ing. Christoph Niethammer High Performance Computing Center Stuttgart (HLRS) Nobelstrasse 19 70569 Stuttgart Tel: ++49(0)711-685-87203 email: christoph.niethammer(a)hlrs.de https://urldefense.us/v3/__https://www.hlrs.de/people/christoph-niethammer_… _______________________________________________ mpi-forum mailing list mpi-forum(a)lists.mpi-forum.org https://urldefense.us/v3/__https://lists.mpi-forum.org/mailman/listinfo/mpi…

1 0

Reading announcement - Clarifying collective synchronisation-semantics
by Christoph Niethammer 16 Feb '26

16 Feb '26

1 0

Last Call for Contributions / MPI-Forum F2F / Request to WG Chairs
by Martin Schulz 16 Feb '26

16 Feb '26

Hi all, Our face-to-face meeting in New York is getting near, and today is the two-week deadline for reading and ballot announcements. If you have any, please send them to the list before the end of the day. Also, if there are any other items for a plenary, please let me know. In general, we are fairly light on official items (kind of expected), and running WG meetings is tricky, as always. On the other hand, we should use the time for some strategic planning towards MPI 6.0: what are the big topics, what is their priority, what should be gating issues, and do we have the right WGs for this? I will also contact all WG chairs directly, but it would be great if all WGs could prepare a discussion on what should be done in the respective WGs for MPI 6.0 and where we should take it technically. If there are any topics outside the existing WG structure, please let me know as well. Thanks! Martin -- Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems TUM School of Computation, Information and Technology (CIT) Dept. of Computer Engineering, TU-Munich, Boltzmannstraße 3, D-85748 Garching Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) Email: schulzm(a)in.tum.de You never fail until you stop trying. (Albert Einstein)

1 0

Fault Tolerance reading/voting announcements
by Joseph Schuchart 16 Feb '26

16 Feb '26

1 0

Reading announcement
by Joseph Schuchart 14 Feb '26

14 Feb '26

1 0

RMA WG reading and plenary announcement
by Joseph Schuchart 13 Feb '26

13 Feb '26

1 0

Re: [Mpi-forum] Issue 467 and I/O working group - Re: Call for In-Person Meeting Topics
by Skjellum, Anthony 12 Feb '26

12 Feb '26

Rolf, We will have to revisit … I do not recall. Tony Anthony Skjellum, PhD 205-807-4968 On Feb 12, 2026, at 11:28 AM, Rolf Rabenseifner via mpi-forum <mpi-forum(a)lists.mpi-forum.org> wrote: External Email Warning This email originated from outside the university. Please use caution when opening attachments, clicking links, or responding to requests. ________________________________ Dear Wes, Tony, Quincey, and Shinji, (same message, but may be with better readable URLs) What is the status of https: //urldefense. us/v3/__https: //github. com/mpi-forum/mpi-issues/issues/467__;!!G_uCfscf7eWS!YlZsRWXWYN8WtAFvschhHy8DpQlo9Dy_iW425PaHLrSGBVC_Ocw4FzMJw1eTgOlz7QJlDjIak6jg84wNmnkFSwbvDrVPyx5vqNRAWg$ ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Dear Wes, Tony, Quincey, and Shinji, (same message, but may be with better readable URLs) What is the status of https://urldefense.us/v3/__https://github.com/mpi-forum/mpi-issues/issues/4… ? What is the status of MPI nonblocking collective shared-filepointer file read and write? I cannot find any open or closed issue # about adding MPI_FILE_IREAD_ORDERED and MPI_FILE_IWRITE_ORDERED. As i remember, it should have existed when MPI_FILE_IREAD_ALL and MPI_IREAD_AT_ALL were added to MPI-3.0. As I commented at https://urldefense.us/v3/__https://github.com/mpi-forum/mpi-issues/issues/4… , the two nonblocking I/O routines MPI_FILE_IREAD_ORDERED and MPI_FILE_IWRITE_ORDERED would be needed before deprecating all split collectives. You may look at See MPI-5.0 Table 14.1 on page 650. I'm only teaching MPI and I want to give correct hints about future directions. Therefore, it would be helpful if the MPI forum and I/O working group can discuss this topic and report the result of that discussion on the issue 467. Best regards Rolf ----- Ursprüngliche Mail ----- > Von: "Main MPI Forum mailing list" <mpi-forum(a)lists.mpi-forum.org> > An: "Main MPI Forum mailing list" <mpi-forum(a)lists.mpi-forum.org> > Gesendet: Freitag, 6. Februar 2026 16:53:54 > Betreff: [Mpi-forum] Call for In-Person Meeting Topics > Hi all, > > If you have any topics for the in-person MPI Forum meeting coming up in a few > weeks, please start announcing them here. Looking for working groups, > plenaries, readings, and votes. > > Also, if you haven’t registered (particularly if you’re planning to attend in > person), please do that here ASAP: https://urldefense.us/v3/__https://www.eventbrite.com/e/1981859425893__;!!G… > > Current registration can be found here: https://urldefense.us/v3/__https://www.mpi-forum.org/meetings/2026/03/atten… > > Thanks, > Wes > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > https://urldefense.us/v3/__https://lists.mpi-forum.org/mailman/listinfo/mpi… -- Dr. Rolf Rabenseifner . . . . . . . . . . rolf.rabenseifner(a)hlrs.de . High Performance Computing Center (HLRS) . . . ++49(0)711/685-65530 . University of Stuttgart . . . . . . https://urldefense.us/v3/__http://www.hlrs.de/people/rabenseifner__;!!G_uCf… . Nobelstr. 19, 70569 Stuttgart, Germany _______________________________________________ mpi-forum mailing list mpi-forum(a)lists.mpi-forum.org https://urldefense.us/v3/__https://lists.mpi-forum.org/mailman/listinfo/mpi…

1 0

Issue 467 and I/O working group - Re: Call for In-Person Meeting Topics
by Rolf Rabenseifner 12 Feb '26

12 Feb '26

1 0