November 2013 - mpi-forum - mailman3.cels.anl.gov

Re: [Mpi-forum] Implementation Status
by Jeff Squyres (jsquyres) 17 Nov '13

17 Nov '13

Rich -- I sent Open MPI updates to Pavan for essentially these same slides yesterday: http://lists.mpi-forum.org/mpi-forum/2013/11/2619.php On Nov 16, 2013, at 10:28 PM, Richard Graham <richardg(a)mellanox.com> wrote: > Attached is the MPI 3.0 implementation status slides I plan to use as part of the MPI Forum BoF. > > Any corrections ? > Rich > <SC_13_mpi_3.0_impl_status.ppt>_______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum -- Jeff Squyres jsquyres(a)cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/

1 0

Implementation Status
by Richard Graham 16 Nov '13

16 Nov '13

Attached is the MPI 3.0 implementation status slides I plan to use as part of the MPI Forum BoF. Any corrections ? Rich

1 0

Re: [Mpi-forum] MPI-3 implementations status
by Pavan Balaji 16 Nov '13

16 Nov '13

Thanks, Jeff. TH-MPI is the MPI on the Tianhe machines in China. I’ll add MS-MPI. — Pavan On Nov 16, 2013, at 5:33 AM, Jeff Squyres (jsquyres) <jsquyres(a)cisco.com> wrote: > Pavan -- > > Just curious -- who is "TH MPI"? Also, I don't see Microsoft MPI on there (unless they're "TH MPI"...?). > > Open MPI updates attached. > > > > > > > On Nov 15, 2013, at 10:14 PM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > > > All, > > > > Attached is the latest MPI implementations slide that is planned to be used at the MPI Forum BoF and other talks at SC. If there are updates to this, please let us know. > > > > Thanks, > > > > — Pavan > > > > > > -- > > Pavan Balaji > > http://www.mcs.anl.gov/~balaji > > > > _______________________________________________ > > mpi-forum mailing list > > mpi-forum(a)lists.mpi-forum.org > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum > > <mpi-implementations.pptx> > > > -- > Jeff Squyres > jsquyres(a)cisco.com > For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ > > <mpi-implementations-4.pptx>_______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Re: [Mpi-forum] MPI-3 implementations status
by Jeff Squyres (jsquyres) 16 Nov '13

16 Nov '13

Pavan -- Just curious -- who is "TH MPI"? Also, I don't see Microsoft MPI on there (unless they're "TH MPI"...?). Open MPI updates attached. On Nov 15, 2013, at 10:14 PM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > All, > > Attached is the latest MPI implementations slide that is planned to be used at the MPI Forum BoF and other talks at SC. If there are updates to this, please let us know. > > Thanks, > > — Pavan > > > -- > Pavan Balaji > http://www.mcs.anl.gov/~balaji > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum > <mpi-implementations.pptx> -- Jeff Squyres jsquyres(a)cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/

1 0

MPI-3 implementations status
by Pavan Balaji 15 Nov '13

15 Nov '13

All, Attached is the latest MPI implementations slide that is planned to be used at the MPI Forum BoF and other talks at SC. If there are updates to this, please let us know. Thanks, — Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Re: [Mpi-forum] Question about the semantics of MPI_Comm_disconnect
by Jeff Squyres (jsquyres) 15 Nov '13

15 Nov '13

On Nov 13, 2013, at 4:25 PM, George Bosilca <bosilca(a)icl.utk.edu> wrote: > Disconnect doesn’t work at the physical connection level, it is supposed to mark two previously connected world as disconnected in the MPI sense. Mmm... not quite. It disconnects two *communicators* -- but the *processes* may still be connected (in the MPI sense). See the example in my previous mail -- you have to call COMM_DISCONNECT twice and FILE_CLOSE to actually disconnect. This is supported by MPI-3 10.5.4 (p399:45-47). > In MPI there is this transitive notion of connected processes, which can only be reversed by the usage of disconnect. Basically MPI_Finalize is collective over all connected processes, which is a burden in some cases. Disconnect is here to allow the disruption of the connected notion. Yeah, I know what disconnect is for. :-) > Thus, Disconnect only has a special meaning when called on a communicator returned by SPAWN, ACCEPT/CONNECT and JOIN. Any other usage is similar to MPI_Comm_Free. This in well explained in 10.5.4 once at the beginning of the section and then again in the first advice to users. I'm not debating what COMM_DISCONNECT actually does -- I'm just saying that it's actually a poor name choice: per 10.5.4, it *doesn't* always disconnect you. Per the example in my previous mail, the processes involved are connected via 3 things: 1. the communicator created by COMM_SPAWN 2. the communicator created by INTERCOMM_MERGE 3. the communicator created by FILE_OPEN You have to COMM_DISCONNECT the first 2, and FILE_CLOSE then 3rd, and *then* you're actually disconnected. Hence, all I'm saying is that "COMM_DISCONNECT" is actually a bit of a misnomer -- the action it's really performing is something more like MPI_COMM_FREE_WAIT (i.e., do a COMM_FREE, but actually wait for all communication to complete on that communicator before returning). For reference, this was my example: >> ----- >> MPI_Comm_spawn(..., &child); >> MPI_Intercomm_merge(child, 0, &child_intra); >> MPI_File_open(child_intra, ..., &file); >> MPI_Comm_disconnect(&child_intra); >> MPI_Comm_disconnect(&child); >> ----- After giving that example, I mused: >> I wonder if this is worth an AtoU... -- Jeff Squyres jsquyres(a)cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/

1 0

Forum readings and votes for tools WG
by Kathryn Mohror 14 Nov '13

14 Nov '13

Hi all, I'm sending this out early so I don't forget due to SC and the holidays. The tools working group has two votes and one reading for the December forum meeting: - Ticket #378 - second vote - Ticket #377 - second vote with minor change - Ticket #400 - reading I've attached a PDF of the tools chapter that has the changes for these tickets in it. Kathryn ______________________________________________________________ Kathryn Mohror, kathryn(a)llnl.gov, http://people.llnl.gov/mohror1 CASC @ Lawrence Livermore National Laboratory, Livermore, CA, USA

1 0

Re: [Mpi-forum] Question about the semantics of MPI_Comm_disconnect
by Rajeev Thakur 13 Nov '13

13 Nov '13

As George said, it seems to be covered by lines 29-40 on pg 399 and 35-40 on 400. Rajeev On Nov 13, 2013, at 3:00 PM, Jeff Squyres (jsquyres) <jsquyres(a)cisco.com> wrote: > On Nov 12, 2013, at 5:40 PM, Rajeev Thakur <thakur(a)mcs.anl.gov> wrote: > >> Think of Comm_disconnect as "whatever connection was there between client and server is gone". > > > Rajeev -- > > This sentence has been bugging me, and I just figured out why. > > I've always thought of COMM_DISCONNECT just as you describe: whatever connection was there between client and server is gone. But what about other types of MPI handles -- like MPI_File and MPI_Window? For example: > > ----- > MPI_Comm_spawn(..., &child); > MPI_Intercomm_merge(child, 0, &child_intra); > MPI_File_open(child_intra, ..., &file); > MPI_Comm_disconnect(&child_intra); > MPI_Comm_disconnect(&child); > ----- > > At this point, all communicators between parent and child are gone, but there's still a File (the same example could be done with an RMA window). > > So the processes are *not* actually disconnected (in the MPI sense of the word "disconnected). Indeed, in the above example, if there was no File, you still have to do *2* disconnects before the parent and child are truly disconnected. > > Hence, I think that this is a very poorly named function -- it *doesn't* necessarily disconnect MPI processes. It really performs and operation that is something like MPI_COMM_FREE_WAIT. > > I wonder if this is worth an AtoU... > > -- > Jeff Squyres > jsquyres(a)cisco.com > For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum

1 0

Re: [Mpi-forum] New Logo for the MPI Forum
by Schulz, Martin 13 Nov '13

13 Nov '13

That would be great - thanks! Martin On Nov 13, 2013, at 2:30 PM, Jerome Vienne <viennej(a)tacc.utexas.edu> wrote: > I can't promise anything due to SC and thanksgiving but I will try to ask to our CMD team to provide me few logos before the meeting. > > Thanks, > Jerome > -- > Jerome Vienne, Ph.D > HPC Software Tools Group > Texas Advanced Computing Center (TACC) > viennej(a)tacc.utexas.edu | Phone: (512) 475-9322 > Office: ROC 1.455B | Fax: (512) 475-9445 > > > On Wed, Nov 13, 2013 at 4:07 PM, Jeff Squyres (jsquyres) <jsquyres(a)cisco.com> wrote: > Yes! We are in dire need of a real logo. > > If you have real artists who can supply some proposals, that would be fabulous. > > > On Nov 13, 2013, at 4:43 PM, Jerome Vienne <viennej(a)tacc.utexas.edu> wrote: > > > Hi Martin, > > > > A "Communication, Media, & Design" team has been created at TACC recently, and I will be happy to work with them to bring few logo proposals before the next meeting. > > Are you still looking for a new logo ? > > > > Thanks, > > Jerome > > -- > > Jerome Vienne, Ph.D > > HPC Software Tools Group > > Texas Advanced Computing Center (TACC) > > viennej(a)tacc.utexas.edu | Phone: (512) 475-9322 > > Office: ROC 1.455B | Fax: (512) 475-9445 > > > > > > On Sat, Sep 14, 2013 at 12:55 PM, Martin Schulz <schulz6(a)llnl.gov> wrote: > > Hi all, > > > > It's long overdue that we use a new logo for the MPI forum - in case you haven't looked at the forum website in a while, it still says MPI-2. At the forum meeting in Madrid we decided on the following process (to make sure we drive this forward). Anyone who has a suggestion for a new logo, please send it to me (until 2 weeks before the meetings). I will collect all of them and then send them to everyone for comments. We will then vote at the next forum meeting in Chicago and pick one. Currently, we have two proposals on the table (created by the UTK folks a while ago - thanks!) - those are attached - but the more choices we have the better. > > > > Get creative … :) > > > > Martin > > > > > > _______________________________________________ > > mpi-forum mailing list > > mpi-forum(a)lists.mpi-forum.org > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum > > > > _______________________________________________ > > mpi-forum mailing list > > mpi-forum(a)lists.mpi-forum.org > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum > > > -- > Jeff Squyres > jsquyres(a)cisco.com > For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/ > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi-forum] New Logo for the MPI Forum
by Jerome Vienne 13 Nov '13

13 Nov '13

I can't promise anything due to SC and thanksgiving but I will try to ask to our CMD team to provide me few logos before the meeting. Thanks, Jerome -- Jerome Vienne, Ph.D HPC Software Tools Group Texas Advanced Computing Center (TACC) viennej(a)tacc.utexas.edu | Phone: (512) 475-9322 Office: ROC 1.455B | Fax: (512) 475-9445 On Wed, Nov 13, 2013 at 4:07 PM, Jeff Squyres (jsquyres) <jsquyres(a)cisco.com > wrote: > Yes! We are in dire need of a real logo. > > If you have real artists who can supply some proposals, that would be > fabulous. > > > On Nov 13, 2013, at 4:43 PM, Jerome Vienne <viennej(a)tacc.utexas.edu> > wrote: > > > Hi Martin, > > > > A "Communication, Media, & Design" team has been created at TACC > recently, and I will be happy to work with them to bring few logo proposals > before the next meeting. > > Are you still looking for a new logo ? > > > > Thanks, > > Jerome > > -- > > Jerome Vienne, Ph.D > > HPC Software Tools Group > > Texas Advanced Computing Center (TACC) > > viennej(a)tacc.utexas.edu | Phone: (512) 475-9322 > > Office: ROC 1.455B | Fax: (512) 475-9445 > > > > > > On Sat, Sep 14, 2013 at 12:55 PM, Martin Schulz <schulz6(a)llnl.gov> > wrote: > > Hi all, > > > > It's long overdue that we use a new logo for the MPI forum - in case you > haven't looked at the forum website in a while, it still says MPI-2. At the > forum meeting in Madrid we decided on the following process (to make sure > we drive this forward). Anyone who has a suggestion for a new logo, please > send it to me (until 2 weeks before the meetings). I will collect all of > them and then send them to everyone for comments. We will then vote at the > next forum meeting in Chicago and pick one. Currently, we have two > proposals on the table (created by the UTK folks a while ago - thanks!) - > those are attached - but the more choices we have the better. > > > > Get creative … :) > > > > Martin > > > > > > _______________________________________________ > > mpi-forum mailing list > > mpi-forum(a)lists.mpi-forum.org > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum > > > > _______________________________________________ > > mpi-forum mailing list > > mpi-forum(a)lists.mpi-forum.org > > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum > > > -- > Jeff Squyres > jsquyres(a)cisco.com > For corporate legal information go to: > http://www.cisco.com/web/about/doing_business/legal/cri/ > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum >

1 0