November 2014 - mpiwg-large-counts - mailman3.cels.anl.gov

Re: [Mpiwg-large-counts] [Mpi-forum] December 2014 Meeting / Agenda
by Jeff Hammond 17 Nov '14

17 Nov '14

1/2 day is fine. Sorry for the delay... Thanks! Jeff On Tue, Nov 4, 2014 at 11:52 PM, Schulz Martin <schulzm(a)llnl.gov> wrote: > Sure, no problem - 1/2 day WG time enough or should be schedule more/less? > > Thanks, > > Martin > > ________________________________________________________________________ > Martin Schulz, schulzm(a)llnl.gov, http://scalability.llnl.gov/ > CASC @ Lawrence Livermore National Laboratory, Livermore, USA > > > > > > On 11/4/14, 8:30 AM, "Jeff Hammond" <jeff.science(a)gmail.com> wrote: > >>Hi Martin, >> >>I would like to have a large-count WG session or two to discuss the >>following tickets: >>- https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/34 and >>https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/338 >>- https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/423 >>- https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/430 >>- (optionally) https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/457 >> >>Background for this request can be found in >>https://github.com/jeffhammond/BigMPI-paper/blob/master/exampi14_resubmiss >>ion_3.pdf?raw=true. >> >>Thanks, >> >>Jeff >> >>On Tue, Nov 4, 2014 at 7:09 AM, Schulz Martin <schulzm(a)llnl.gov> wrote: >>> Hi all, >>> >>> I put up an initial roster for the December meeting in San Jose. As >>>usual, I >>> am planning to arrange WG meeting in the beginning of the meeting (until >>> Tuesday noon or evening depending on the requests that I get), followed >>>by >>> the plenaries. Votes will be Thursday morning. >>> >>> For the plenaries, I would like to try something different, based on the >>> experience at the forum in Japan: instead of having a fixed schedule, >>>I¹ll >>> put up a list of topics that we need to cover and we¹ll go through them >>>in >>> order, but not necessarily with a fixed schedule. We basically followed >>>this >>> scheme in Japan (out of necessity since we had to redo the agenda on >>>the fly >>> due to the missed quorum) and this worked fairly well and allowed for >>>enough >>> discussions for the topics that needed them. I¹ll still try to keep >>>track of >>> overall time to make sure we get through with the agenda, in particular >>>with >>> the items to be voted on, but I would like to avoid that this dictates >>>the >>> schedule. >>> >>> For the WG leads: please send me any requests for WG time or plenary >>>items >>> as soon as you can (even if you need to revise that later), so we can >>>start >>> building a more concrete agenda. Also, please let me know of any tickets >>> that you want read or voted on as part of errata. I have pre-populated >>>the >>> agenda with the second and first votes, since those are fixed but >>>please >>> also double check those. The deadline for announcing tickets for >>>readings or >>> votes is as usual 2 weeks before the forum, which is the Monday 11/24, >>>I.e., >>> the Monday of Thanksgiving week. >>> >>> Thanks! >>> >>> Martin >>> >>> >>> >>> ________________________________________________________________________ >>> Martin Schulz, schulzm(a)llnl.gov, http://scalability.llnl.gov/ >>> 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 >> >> >> >>-- >>Jeff Hammond >>jeff.science(a)gmail.com >>http://jeffhammond.github.io/ > > -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0

Re: [Mpiwg-large-counts] [Mpi-forum] December 2014 Meeting / Agenda
by Schulz Martin 04 Nov '14

04 Nov '14

Sure, no problem - 1/2 day WG time enough or should be schedule more/less? Thanks, Martin ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://scalability.llnl.gov/ CASC @ Lawrence Livermore National Laboratory, Livermore, USA On 11/4/14, 8:30 AM, "Jeff Hammond" <jeff.science(a)gmail.com> wrote: >Hi Martin, > >I would like to have a large-count WG session or two to discuss the >following tickets: >- https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/34 and >https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/338 >- https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/423 >- https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/430 >- (optionally) https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/457 > >Background for this request can be found in >https://github.com/jeffhammond/BigMPI-paper/blob/master/exampi14_resubmiss >ion_3.pdf?raw=true. > >Thanks, > >Jeff > >On Tue, Nov 4, 2014 at 7:09 AM, Schulz Martin <schulzm(a)llnl.gov> wrote: >> Hi all, >> >> I put up an initial roster for the December meeting in San Jose. As >>usual, I >> am planning to arrange WG meeting in the beginning of the meeting (until >> Tuesday noon or evening depending on the requests that I get), followed >>by >> the plenaries. Votes will be Thursday morning. >> >> For the plenaries, I would like to try something different, based on the >> experience at the forum in Japan: instead of having a fixed schedule, >>I¹ll >> put up a list of topics that we need to cover and we¹ll go through them >>in >> order, but not necessarily with a fixed schedule. We basically followed >>this >> scheme in Japan (out of necessity since we had to redo the agenda on >>the fly >> due to the missed quorum) and this worked fairly well and allowed for >>enough >> discussions for the topics that needed them. I¹ll still try to keep >>track of >> overall time to make sure we get through with the agenda, in particular >>with >> the items to be voted on, but I would like to avoid that this dictates >>the >> schedule. >> >> For the WG leads: please send me any requests for WG time or plenary >>items >> as soon as you can (even if you need to revise that later), so we can >>start >> building a more concrete agenda. Also, please let me know of any tickets >> that you want read or voted on as part of errata. I have pre-populated >>the >> agenda with the second and first votes, since those are fixed but >>please >> also double check those. The deadline for announcing tickets for >>readings or >> votes is as usual 2 weeks before the forum, which is the Monday 11/24, >>I.e., >> the Monday of Thanksgiving week. >> >> Thanks! >> >> Martin >> >> >> >> ________________________________________________________________________ >> Martin Schulz, schulzm(a)llnl.gov, http://scalability.llnl.gov/ >> 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 > > > >-- >Jeff Hammond >jeff.science(a)gmail.com >http://jeffhammond.github.io/

1 0

Re: [Mpiwg-large-counts] [Mpi-forum] December 2014 Meeting / Agenda
by Jeff Hammond 04 Nov '14

04 Nov '14

Hi Martin, I would like to have a large-count WG session or two to discuss the following tickets: - https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/34 and https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/338 - https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/423 - https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/430 - (optionally) https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/457 Background for this request can be found in https://github.com/jeffhammond/BigMPI-paper/blob/master/exampi14_resubmissi…. Thanks, Jeff On Tue, Nov 4, 2014 at 7:09 AM, Schulz Martin <schulzm(a)llnl.gov> wrote: > Hi all, > > I put up an initial roster for the December meeting in San Jose. As usual, I > am planning to arrange WG meeting in the beginning of the meeting (until > Tuesday noon or evening depending on the requests that I get), followed by > the plenaries. Votes will be Thursday morning. > > For the plenaries, I would like to try something different, based on the > experience at the forum in Japan: instead of having a fixed schedule, I’ll > put up a list of topics that we need to cover and we’ll go through them in > order, but not necessarily with a fixed schedule. We basically followed this > scheme in Japan (out of necessity since we had to redo the agenda on the fly > due to the missed quorum) and this worked fairly well and allowed for enough > discussions for the topics that needed them. I’ll still try to keep track of > overall time to make sure we get through with the agenda, in particular with > the items to be voted on, but I would like to avoid that this dictates the > schedule. > > For the WG leads: please send me any requests for WG time or plenary items > as soon as you can (even if you need to revise that later), so we can start > building a more concrete agenda. Also, please let me know of any tickets > that you want read or voted on as part of errata. I have pre-populated the > agenda with the second and first votes, since those are fixed – but please > also double check those. The deadline for announcing tickets for readings or > votes is as usual 2 weeks before the forum, which is the Monday 11/24, I.e., > the Monday of Thanksgiving week. > > Thanks! > > Martin > > > > ________________________________________________________________________ > Martin Schulz, schulzm(a)llnl.gov, http://scalability.llnl.gov/ > 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 -- Jeff Hammond jeff.science(a)gmail.com http://jeffhammond.github.io/

1 0