- mpiwg-io - mailman3.cels.anl.gov

Re: [MPI3-IO] Topics for discussion for the I/O working group slot
by Rob Latham 26 Feb '13

26 Feb '13

On Mon, Feb 25, 2013 at 05:26:28PM -0600, Mohamad Chaarawi wrote: > Hi Dries, > > On 2/25/2013 3:44 PM, Dries Kimpe wrote: > >- Fix shared file pointer access cfr. the EuroMPI paper? > > yes thanks for reminding me :-) > > >- independent file view changes > > Do you think, implementation wise, that it would be straightforward > to just remove this restriction? I'm worried how that would effect > shared file pointer operations in the standard. > I'm all for having a new routine to change the file view > independently with restrictions on what comes after that. Yes, please. MPI_File_set_view_not_dumb would be a fine name. Or one could introduce a crapton of routines that take a file view in addition to a memory datatype and obviate the need for setting file view at all (well except for data representation but who actually does that?) ==rob -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA

1 0

Re: [MPI3-IO] Topics for discussion for the I/O working group slot
by Mohamad Chaarawi 25 Feb '13

25 Feb '13

Hi Quincey/Dries, On 2/25/2013 5:09 PM, Dries Kimpe wrote: > * Quincey Koziol <koziol(a)hdfgroup.org> [2013-02-25 16:08:47]: > >> I was thinking more about this, and what about the "group collective" >> I/O operations that we've kicked around a bit? (Where "group >> collective" is a sub-group of the processes that opened the file) > Or the alternative: a subgroup-open which can quickly generate a handle > for a subgroup given the handle for the larger group. Yes i like that better than copying and modifying the entire I/O operations. I don't imagine the sub-group open is a real file-system open (right?) so it should be lightweight? Thanks, Mohamad > > Dries > > > _______________________________________________ > MPI3-IO mailing list > MPI3-IO(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io

1 0

Re: [MPI3-IO] Topics for discussion for the I/O working group slot
by Quincey Koziol 25 Feb '13

25 Feb '13

On Feb 25, 2013, at 5:09 PM, Dries Kimpe <dkimpe(a)mcs.anl.gov> wrote: > > * Quincey Koziol <koziol(a)hdfgroup.org> [2013-02-25 16:08:47]: > >> I was thinking more about this, and what about the "group collective" >> I/O operations that we've kicked around a bit? (Where "group >> collective" is a sub-group of the processes that opened the file) > > Or the alternative: a subgroup-open which can quickly generate a handle > for a subgroup given the handle for the larger group. This would be fine, as long as it was fast - I would expect that the subgroup-opened file handle would only exist for a single HDF5 I/O call (read or write). Quincey

1 0

Re: [MPI3-IO] Topics for discussion for the I/O working group slot
by Mohamad Chaarawi 25 Feb '13

25 Feb '13

Hi Dries, On 2/25/2013 3:44 PM, Dries Kimpe wrote: > - Fix shared file pointer access cfr. the EuroMPI paper? yes thanks for reminding me :-) > - independent file view changes Do you think, implementation wise, that it would be straightforward to just remove this restriction? I'm worried how that would effect shared file pointer operations in the standard. I'm all for having a new routine to change the file view independently with restrictions on what comes after that. Thanks, Mohamad

1 0

Re: [MPI3-IO] Topics for discussion for the I/O working group slot
by Dries Kimpe 25 Feb '13

25 Feb '13

* Quincey Koziol <koziol(a)hdfgroup.org> [2013-02-25 16:08:47]: > I was thinking more about this, and what about the "group collective" > I/O operations that we've kicked around a bit? (Where "group > collective" is a sub-group of the processes that opened the file) Or the alternative: a subgroup-open which can quickly generate a handle for a subgroup given the handle for the larger group. Dries

1 0

Re: [MPI3-IO] Topics for discussion for the I/O working group slot
by Quincey Koziol 25 Feb '13

25 Feb '13

I was thinking more about this, and what about the "group collective" I/O operations that we've kicked around a bit? (Where "group collective" is a sub-group of the processes that opened the file) Quincey On Feb 25, 2013, at 3:44 PM, Dries Kimpe <dkimpe(a)mcs.anl.gov> wrote: > - Fix shared file pointer access cfr. the EuroMPI paper? > - independent file view changes > > > Dries > > * Mohamad Chaarawi <chaarawi(a)hdfgroup.org> [2013-02-25 15:41:16]: > >> Hi All, > >> The I/O working group has 1hr and 45 minutes of time allocated at the Forum >> meeting in a couple of weeks. >> HDF doesn't have a lot of things to talk about, other than current status of >> the non-blocking collective I/O ticket and maybe a non-blocking open/close.. >> But I don't imagine those would take more than 1/2 an hr. >> Does anyone have anything they would want to discuss then? > >> Thanks, >> Mohamad >> _______________________________________________ >> MPI3-IO mailing list >> MPI3-IO(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io > _______________________________________________ > MPI3-IO mailing list > MPI3-IO(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io

1 0

Re: [MPI3-IO] Topics for discussion for the I/O working group slot
by Dries Kimpe 25 Feb '13

25 Feb '13

- Fix shared file pointer access cfr. the EuroMPI paper? - independent file view changes Dries * Mohamad Chaarawi <chaarawi(a)hdfgroup.org> [2013-02-25 15:41:16]: > Hi All, > The I/O working group has 1hr and 45 minutes of time allocated at the Forum > meeting in a couple of weeks. > HDF doesn't have a lot of things to talk about, other than current status of > the non-blocking collective I/O ticket and maybe a non-blocking open/close.. > But I don't imagine those would take more than 1/2 an hr. > Does anyone have anything they would want to discuss then? > Thanks, > Mohamad > _______________________________________________ > MPI3-IO mailing list > MPI3-IO(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io

1 0

[MPI3-IO] Topics for discussion for the I/O working group slot
by Mohamad Chaarawi 25 Feb '13

25 Feb '13

Hi All, The I/O working group has 1hr and 45 minutes of time allocated at the Forum meeting in a couple of weeks. HDF doesn't have a lot of things to talk about, other than current status of the non-blocking collective I/O ticket and maybe a non-blocking open/close.. But I don't imagine those would take more than 1/2 an hr. Does anyone have anything they would want to discuss then? Thanks, Mohamad

1 0

Re: [MPI3-IO] I/O chapter
by Rajeev Thakur 02 Jul '12

02 Jul '12

I have applied ticket 140. I am assuming nothing else needs to be done to the I/O chapter. Rajeev On Jun 24, 2012, at 5:44 PM, Fab Tillier wrote: > The latest for that ticket is in SVN: > URL: https://svn.mpi-forum.org/svn/mpi-forum-docs/trunk/working-groups/mpi-3/bw-… > > Easiest would likely be to do an svn merge. > > Cheers, > -Fab > > -----Original Message----- > From: mpi3-io-bounces(a)lists.mpi-forum.org [mailto:[email protected]] On Behalf Of Rajeev Thakur > Sent: Sunday, June 24, 2012 3:04 PM > To: Discussion about I/O ideas for future MPI standards > Cc: rabenseifner(a)hlrs.de Rabenseifner; Torsten(a)osl.iu.edu; Hoefler; Richard Graham > Subject: Re: [MPI3-IO] I/O chapter > > Should I apply the patch jeff-140.patch from the ticket page? > https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/140 > > Rajeev > > On Jun 24, 2012, at 2:36 PM, Fab Tillier wrote: > >> Ticket 140 (const) touched the I/O chapter. Should be pretty easy to merge, though. >> >> Cheers, >> -Fab >> >> -----Original Message----- >> From: mpi3-io-bounces(a)lists.mpi-forum.org [mailto:[email protected]] On Behalf Of Rajeev Thakur >> Sent: Sunday, June 24, 2012 7:20 AM >> To: Discussion about I/O ideas for future MPI standards >> Cc: rabenseifner(a)hlrs.de Rabenseifner; Torsten Hoefler; Richard Graham >> Subject: [MPI3-IO] I/O chapter >> >> Other than a few bug fixes in the examples, which Bill has already checked in, is there anything that needs to be merged into the I/O chapter? I think not since the nonblocking collective file I/O functions didn't make it. If there is something to be checked in let me know. A two-hour plenary slot on Thursday morning shouldn't be needed for the I/O chapter. >> >> Rajeev >> >> >> _______________________________________________ >> MPI3-IO mailing list >> MPI3-IO(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io >> >> >> >> _______________________________________________ >> MPI3-IO mailing list >> MPI3-IO(a)lists.mpi-forum.org >> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io > > > _______________________________________________ > MPI3-IO mailing list > MPI3-IO(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io > > > > _______________________________________________ > MPI3-IO mailing list > MPI3-IO(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io

1 0

Re: [MPI3-IO] I/O chapter
by Fab Tillier 24 Jun '12

24 Jun '12

The latest for that ticket is in SVN: URL: https://svn.mpi-forum.org/svn/mpi-forum-docs/trunk/working-groups/mpi-3/bw-… Easiest would likely be to do an svn merge. Cheers, -Fab -----Original Message----- From: mpi3-io-bounces(a)lists.mpi-forum.org [mailto:[email protected]] On Behalf Of Rajeev Thakur Sent: Sunday, June 24, 2012 3:04 PM To: Discussion about I/O ideas for future MPI standards Cc: rabenseifner(a)hlrs.de Rabenseifner; Torsten(a)osl.iu.edu; Hoefler; Richard Graham Subject: Re: [MPI3-IO] I/O chapter Should I apply the patch jeff-140.patch from the ticket page? https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/140 Rajeev On Jun 24, 2012, at 2:36 PM, Fab Tillier wrote: > Ticket 140 (const) touched the I/O chapter. Should be pretty easy to merge, though. > > Cheers, > -Fab > > -----Original Message----- > From: mpi3-io-bounces(a)lists.mpi-forum.org [mailto:[email protected]] On Behalf Of Rajeev Thakur > Sent: Sunday, June 24, 2012 7:20 AM > To: Discussion about I/O ideas for future MPI standards > Cc: rabenseifner(a)hlrs.de Rabenseifner; Torsten Hoefler; Richard Graham > Subject: [MPI3-IO] I/O chapter > > Other than a few bug fixes in the examples, which Bill has already checked in, is there anything that needs to be merged into the I/O chapter? I think not since the nonblocking collective file I/O functions didn't make it. If there is something to be checked in let me know. A two-hour plenary slot on Thursday morning shouldn't be needed for the I/O chapter. > > Rajeev > > > _______________________________________________ > MPI3-IO mailing list > MPI3-IO(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io > > > > _______________________________________________ > MPI3-IO mailing list > MPI3-IO(a)lists.mpi-forum.org > http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io _______________________________________________ MPI3-IO mailing list MPI3-IO(a)lists.mpi-forum.org http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io

1 0