March 2014 - mpi-forum - mailman3.cels.anl.gov

Re: [Mpi-forum] info key text normative?
by Fab Tillier 13 Mar '14

13 Mar '14

Hi Jeff, Jeff Hammond wrote on Thu, 13 Mar 2014 at 09:48:59 > Is the text that describes info keys normative, i.e. binding? Yes, the mechanics of info keys are well defined, and the standard defines reserved keys for file objects (page 500). > I thought implementations were free to ignore info keys and thus > the text defining them was not necessarily normative. Implementations can ignore keys, but this doesn't change how info keys behave from an interface perspective. At least that's my take on it. -Fab

1 0

MPI_ERR_UNSUPPORTED_DATAREP error class definition scoped too tightly
by Fab Tillier 13 Mar '14

13 Mar '14

Hi folks, The standard defines MPI_ERR_UNSUPPORTED_DATAREP as an I/O error class, with the meaning "Unsupported datarep passed to MPI_FILE_SET_VIEW". However, datareps are also passed as input to MPI_PACK_EXTERNAL and MPI_UNPACK_EXTERNAL. Shouldn't these functions use the same error class for unsupported datareps, rather than MPI_ERR_ARG? Can we update the description of MPI_ERR_UNSUPPORTED_DATAREP (page 551, line 10) to also include MPI_PACK_EXTERNAL and MPI_UNPACK_EXTERNAL as a ticket-0 item for MPI 3.1, or will this need to happen in a later version? Thanks, -Fab

1 0

info key text normative?
by Jeff Hammond 13 Mar '14

13 Mar '14

Is the text that describes info keys normative, i.e. binding? I thought implementations were free to ignore info keys and thus the text defining them was not necessarily normative. Thanks, Jeff -- Jeff Hammond jeff.science(a)gmail.com

1 0

Re: [Mpi-forum] Timeline for MPI 3.1
by William Gropp 07 Mar '14

07 Mar '14

Rolf, The RMA chapter committee will be adding this example. I'm not convinced that the suggested location is the correct place, and will be looking into it more. As an example to illuminate existing functionality, it isn't covered by the "technical additions" requirement. Bill William Gropp Director, Parallel Computing Institute Thomas M. Siebel Chair in Computer Science University of Illinois Urbana-Champaign On Mar 7, 2014, at 9:53 AM, Rolf Rabenseifner wrote: > Dear Martin and Bill, > > I would like to see #413 in MPI-3.1. > Bill, you wanted to manage the progress on this ticket. > I'm already teaching the content and must be sure > that it is correct. > I do not know, whether the RMA group needs > "one (or more) rounds of opportunities". > > Best regards > Rolf > > ----- Original Message ----- >> From: "Martin Schulz" <schulzm(a)llnl.gov> >> To: "Main MPI Forum mailing list" >> Sent: Thursday, March 6, 2014 2:41:42 AM >> Subject: [Mpi-forum] Timeline for MPI 3.1 >> >> Hi all, >> >> As discussed during the last meeting, as well as earlier this week, >> we'll make a decision tomorrow at the MPI forum about the time line >> for MPI 3.1. The main decision is whether we freeze the technical >> additions to MPI 3.1 or whether there are significant open issues >> (within the scope of MPI 3.1, not major additions suitable for MPI >> 4.0) for which we need one (or more) rounds of opportunities for >> readings. If you know of any issue that you think falls into the >> latter category, please send email to the list or to myself. In >> particular, it would be great if WG chairs could look over the open >> issues in their WG and let me know. >> >> Thanks! >> >> Martin >> >> >> ________________________________________________________________________ >> Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm >> CASC @ Lawrence Livermore National Laboratory, Livermore, USA >> >> >> >> _______________________________________________ >> mpi-forum mailing list >> mpi-forum(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum >> > > -- > Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)hlrs.de > High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 > University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 > Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner > Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307) > _______________________________________________ > 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] Timeline for MPI 3.1
by Rolf Rabenseifner 07 Mar '14

07 Mar '14

Dear Martin and Bill, I would like to see #413 in MPI-3.1. Bill, you wanted to manage the progress on this ticket. I'm already teaching the content and must be sure that it is correct. I do not know, whether the RMA group needs "one (or more) rounds of opportunities". Best regards Rolf ----- Original Message ----- > From: "Martin Schulz" <schulzm(a)llnl.gov> > To: "Main MPI Forum mailing list" > Sent: Thursday, March 6, 2014 2:41:42 AM > Subject: [Mpi-forum] Timeline for MPI 3.1 > > Hi all, > > As discussed during the last meeting, as well as earlier this week, > we'll make a decision tomorrow at the MPI forum about the time line > for MPI 3.1. The main decision is whether we freeze the technical > additions to MPI 3.1 or whether there are significant open issues > (within the scope of MPI 3.1, not major additions suitable for MPI > 4.0) for which we need one (or more) rounds of opportunities for > readings. If you know of any issue that you think falls into the > latter category, please send email to the list or to myself. In > particular, it would be great if WG chairs could look over the open > issues in their WG and let me know. > > Thanks! > > Martin > > > ________________________________________________________________________ > Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm > CASC @ Lawrence Livermore National Laboratory, Livermore, USA > > > > _______________________________________________ > mpi-forum mailing list > mpi-forum(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi-forum > -- Dr. Rolf Rabenseifner . . . . . . . . . .. email rabenseifner(a)hlrs.de High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530 University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832 Head of Dpmt Parallel Computing . . . www.hlrs.de/people/rabenseifner Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307)

1 0

Re: [Mpi-forum] Slides from the meeting
by Todd Gamblin 07 Mar '14

07 Mar '14

I think you have this already, but here it is again. This is the github repo I presented from: https://github.com/tgamblin/mpi-resilience On Mar 6, 2014, at 10:58 AM, Jeff Squyres (jsquyres) <jsquyres(a)cisco.com> wrote: > Please send me your slides that were presented this week so that I can put them on the web page. > > Thanks. > > -- > 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 ______________________________________________________________________ Todd Gamblin, tgamblin(a)llnl.gov, http://people.llnl.gov/gamblin2 CASC @ Lawrence Livermore National Laboratory, Livermore, CA, USA

1 0

Slides from the meeting
by Jeff Squyres (jsquyres) 06 Mar '14

06 Mar '14

Please send me your slides that were presented this week so that I can put them on the web page. Thanks. -- Jeff Squyres jsquyres(a)cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/

1 0

Timeline for MPI 3.1
by Schulz, Martin 05 Mar '14

05 Mar '14

Hi all, As discussed during the last meeting, as well as earlier this week, we'll make a decision tomorrow at the MPI forum about the time line for MPI 3.1. The main decision is whether we freeze the technical additions to MPI 3.1 or whether there are significant open issues (within the scope of MPI 3.1, not major additions suitable for MPI 4.0) for which we need one (or more) rounds of opportunities for readings. If you know of any issue that you think falls into the latter category, please send email to the list or to myself. In particular, it would be great if WG chairs could look over the open issues in their WG and let me know. Thanks! Martin ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Straw man Reinit interface
by Todd Gamblin 03 Mar '14

03 Mar '14

Here's the straw man interface I'm going to present: https://github.com/tgamblin/mpi-resilience ______________________________________________________________________ Todd Gamblin, tgamblin(a)llnl.gov, http://people.llnl.gov/gamblin2 CASC @ Lawrence Livermore National Laboratory, Livermore, CA, USA

1 0