August 2010 - mpiwg-tools - mailman3.cels.anl.gov

Re: [Mpi3-tools] Final draft of MPIR Document
by Jeff Squyres 30 Aug '10

30 Aug '10

On Aug 27, 2010, at 7:45 PM, Bronis R. de Supinski wrote: > I'll refraing from other grammatical suggestions. Please don't! Your grammatical suggestions are frequently very good... -- Jeff Squyres jsquyres(a)cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/

1 0

Re: [Mpi3-tools] Final draft of MPIR Document
by Martin Schulz 27 Aug '10

27 Aug '10

On Aug 27, 2010, at 5:05 PM, Dong H. Ahn wrote: > Hi Martin, > > Whatever term you decided to use, please use it consistently across > the > text. There are several places where the term, MPIR launch/attach > model, > is used in the text. Thanks for catching this - fixed and checked into SVN. Martin > > Thanks. > > Dong > > On 8/27/2010 4:55 PM, Martin Schulz wrote: >> On Aug 27, 2010, at 4:45 PM, Bronis R. de Supinski wrote: >> >>> Martin: >>> >>> I'm not sure why the term "modi" was used but in the >>> least it is a rather arcane usage. I see nothing in >>> the Wikipedia list that is applicable: >> I actually added that in the last round and I can certainly switch it >> to modes (I was debating that) - originally we had models there, >> but that didn't seem right to me. Other suggestions are welcome. >> >> Martin >> >> >>> http://***en.wikipedia.org/wiki/Modi >>> >>> I suggest you change this to "modes"... >>> >>> I'll refraing from other grammatical suggestions. >>> >>> Bronis >>> >>> >>> >>> On Fri, 27 Aug 2010, Martin Schulz wrote: >>> >>>> Hi all, >>>> >>>> Attached is the final version of the MPIR document, as we are >>>> planning >>>> to present it to >>>> the MPI forum. Please speak up if you have any final comments or >>>> concerns. Otherwise, >>>> we are planning to submit this version on Monday to the forum for >>>> review for the >>>> September meeting. >>>> >>>> Martin >>>> >>>> >>> _______________________________________________ >>> Mpi3-tools mailing list >>> Mpi3-tools(a)lists.mpi-forum.org >>> http://***lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools >>> >> ________________________________________________________________________ >> Martin Schulz, schulzm(a)llnl.gov, http://**people.llnl.gov/schulzm >> CASC @ Lawrence Livermore National Laboratory, Livermore, USA >> >> >> >> _______________________________________________ >> Mpi3-tools mailing list >> Mpi3-tools(a)lists.mpi-forum.org >> http://**lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools >> > > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://*lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools > ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] Final draft of MPIR Document
by Dong H. Ahn 27 Aug '10

27 Aug '10

Hi Martin, Whatever term you decided to use, please use it consistently across the text. There are several places where the term, MPIR launch/attach model, is used in the text. Thanks. Dong On 8/27/2010 4:55 PM, Martin Schulz wrote: > On Aug 27, 2010, at 4:45 PM, Bronis R. de Supinski wrote: > >> Martin: >> >> I'm not sure why the term "modi" was used but in the >> least it is a rather arcane usage. I see nothing in >> the Wikipedia list that is applicable: > I actually added that in the last round and I can certainly switch it > to modes (I was debating that) - originally we had models there, > but that didn't seem right to me. Other suggestions are welcome. > > Martin > > >> http://**en.wikipedia.org/wiki/Modi >> >> I suggest you change this to "modes"... >> >> I'll refraing from other grammatical suggestions. >> >> Bronis >> >> >> >> On Fri, 27 Aug 2010, Martin Schulz wrote: >> >>> Hi all, >>> >>> Attached is the final version of the MPIR document, as we are >>> planning >>> to present it to >>> the MPI forum. Please speak up if you have any final comments or >>> concerns. Otherwise, >>> we are planning to submit this version on Monday to the forum for >>> review for the >>> September meeting. >>> >>> Martin >>> >>> >> _______________________________________________ >> Mpi3-tools mailing list >> Mpi3-tools(a)lists.mpi-forum.org >> http://**lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools >> > ________________________________________________________________________ > Martin Schulz, schulzm(a)llnl.gov, http://*people.llnl.gov/schulzm > CASC @ Lawrence Livermore National Laboratory, Livermore, USA > > > > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://*lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools >

1 0

Re: [Mpi3-tools] Final draft of MPIR Document
by Martin Schulz 27 Aug '10

27 Aug '10

On Aug 27, 2010, at 4:45 PM, Bronis R. de Supinski wrote: > > Martin: > > I'm not sure why the term "modi" was used but in the > least it is a rather arcane usage. I see nothing in > the Wikipedia list that is applicable: I actually added that in the last round and I can certainly switch it to modes (I was debating that) - originally we had models there, but that didn't seem right to me. Other suggestions are welcome. Martin > > http://*en.wikipedia.org/wiki/Modi > > I suggest you change this to "modes"... > > I'll refraing from other grammatical suggestions. > > Bronis > > > > On Fri, 27 Aug 2010, Martin Schulz wrote: > >> Hi all, >> >> Attached is the final version of the MPIR document, as we are >> planning >> to present it to >> the MPI forum. Please speak up if you have any final comments or >> concerns. Otherwise, >> we are planning to submit this version on Monday to the forum for >> review for the >> September meeting. >> >> Martin >> >> > _______________________________________________ > Mpi3-tools mailing list > Mpi3-tools(a)lists.mpi-forum.org > http://*lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools > ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] Final draft of MPIR Document
by Bronis R. de Supinski 27 Aug '10

27 Aug '10

Martin: I'm not sure why the term "modi" was used but in the least it is a rather arcane usage. I see nothing in the Wikipedia list that is applicable: http://en.wikipedia.org/wiki/Modi I suggest you change this to "modes"... I'll refraing from other grammatical suggestions. Bronis On Fri, 27 Aug 2010, Martin Schulz wrote: > Hi all, > > Attached is the final version of the MPIR document, as we are planning > to present it to > the MPI forum. Please speak up if you have any final comments or > concerns. Otherwise, > we are planning to submit this version on Monday to the forum for > review for the > September meeting. > > Martin > >

1 0

[Mpi3-tools] Final draft of MPIR Document
by Martin Schulz 27 Aug '10

27 Aug '10

Hi all, Attached is the final version of the MPIR document, as we are planning to present it to the MPI forum. Please speak up if you have any final comments or concerns. Otherwise, we are planning to submit this version on Monday to the forum for review for the September meeting. Martin ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] Next Tools WG Meeting 8/23 - MPIR & MPIT
by Martin Schulz 23 Aug '10

23 Aug '10

Hi all, Since we had to end quickly today, just a few organizational things via email: - The notes from today's meeting are online - please feel free to add if necessary. - We didn't cover the last part of the MPIT discussion, so please send any comments via email (preferably this week) so that I can create a new draft of the MPIT document - We will need to get the MPIR document to the MPI forum by the end of the week, hence if there are any comments left, please send them asap - The meeting in two weeks is canceled due to Labor day - The next meeting is on 9/20 at the usual time - the agenda will be to cover any open issues on MPIR raised during the forum meeting the week before and to start going over the new MPIT draft. Martin On Aug 22, 2010, at 2:58 PM, Martin Schulz wrote: > Hi all, > > Our next tools WG meeting is (as usual) on Monday 8/23 (tomorrow) at > 8am PDT, > 11am EDT, 5pm MSZ. Jeff created a Webex session for us, which is > linked on the > tools Wiki. It is: > > https://cisco.webex.com/ciscosales/j.php?ED=145066572&UID=0&PW=NMjcyNzAwZmU… > Password: tools > > We have two main item for the agenda: MPIR and MPIT. > > MPIR: John requested some time to finish the discussion on the MPIR > document before > we turn the document into a standard compliant Latex document and > give it for to the > MPI forum forum for review (has to happen by the end of next week). > > MPIT: I recently gave a presentation on MPIT to the tools community > and I would like to > discuss the feedback from that meeting and its impact on the MPIT > proposal. More detail > on this are below. > > Talk to you tomorrow / Monday, > > Martin > > > > Feedback from the tools community on MPIT (based on CScADS > discussions): > (for those of you who were at the CScADS workshop, please feel free > to add) > > Collection of use cases > Current ideas in MPIT seem to match those use cases very well > High/Low watermarks seen as quite useful > Identification of exhausted resources important > > Naming > Several name changes suggested > configuration variables -> control variables (as in OpenMP) > Prefixes: > MPIT_CTRLVARS_... > MPIT_PERFVARS_... > > Verbosity levels > Why restrict verbosity from 1 to 5? > Allow a query for the max. verbosity N and then allow levels 1-N > > Initialization > Suggestion to remove IsInitialized and IsFinalized > (not useful if we can initialize MPIT multiple times) > > Bias towards adding Fortran routines > Need more feedback from the Fortran gurus in the MPI forum > > Control variables > Very strong wish to allow the setting of control variables > In particular autotuning community is very interested and promised to > write up several concrete use cases > How to deal with setting the variables from multiple tools > a) provide lock/unlock routines > b) allow one tool to Initialize MPIT exclusively in MPIT_Init > c) Do nothing and leave races up to tools/users > How to deal with synchronizing variables > Provide additional information during query: > readonly - can never be set > sync/nosync - shows whether a variables is synchronizing or not > communicator - setting a synchronizing variables requires a global > operation on this communicator > (can be MPI_COMM_SELF for nosync variables) > Set routine takes communicator as argument > Must be called by all members of the communicator > > Extensibility > Very strong wish to have the option to extend this interface > Tools should have the ability to provide new variables > (this was brought up by several tools groups for various needs) > > Seen as essential (not just for Extensibility): Profiling routines > for MPIT > > Would like to see a different iterator concept: > change from iterators to a list of variables with an integer index > (this fixes the type of the iterator) > Query the number of variables and then provide query routines to ask > for more information on each variable > Restrictions: variables can not be removed or reordered, just added > > How to return query information: > Return a struct instead of a list of arguments > Cleaner ways to integrate new information in future versions > Need separate MPIT version constant to guard extensions > > Ideally: would like to give MPI an address for a variable to watch > and MPI > would then integrate this variable into the MPIT variables > -> easy extension for new variables by external tools through same > API > > Alternatively: tools can hide this in an PMPIT layer, but then need > additional > mechanisms to allocate "dummy" handles to avoid duplicating the > handle space > > > > > > > > ________________________________________________________________________ > Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm > CASC @ Lawrence Livermore National Laboratory, Livermore, USA > > > ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] Next Tools WG Meeting 8/23 - MPIR & MPIT
by Bronis R. de Supinski 22 Aug '10

22 Aug '10

Martin: One correction -- the OpenMP term is "internal control variables" or "ICVs". Bronis On Sun, 22 Aug 2010, Martin Schulz wrote: > Hi all, > > Our next tools WG meeting is (as usual) on Monday 8/23 (tomorrow) at 8am PDT, > 11am EDT, 5pm MSZ. Jeff created a Webex session for us, which is linked on the > tools Wiki. It is: > > https://*cisco.webex.com/ciscosales/j.php?ED=145066572&UID=0&PW=NMjcyNzAwZmU2&RT=MiMxMQ%3D%3D > Password: tools > > We have two main item for the agenda: MPIR and MPIT. > > MPIR: John requested some time to finish the discussion on the MPIR document before > we turn the document into a standard compliant Latex document and give it for to the > MPI forum forum for review (has to happen by the end of next week). > > MPIT: I recently gave a presentation on MPIT to the tools community and I would like to > discuss the feedback from that meeting and its impact on the MPIT proposal. More detail > on this are below. > > Talk to you tomorrow / Monday, > > Martin > > > > Feedback from the tools community on MPIT (based on CScADS discussions): > (for those of you who were at the CScADS workshop, please feel free to add) > > Collection of use cases > Current ideas in MPIT seem to match those use cases very well > High/Low watermarks seen as quite useful > Identification of exhausted resources important > > Naming > Several name changes suggested > configuration variables -> control variables (as in OpenMP) > Prefixes: > MPIT_CTRLVARS_... > MPIT_PERFVARS_... > > Verbosity levels > Why restrict verbosity from 1 to 5? > Allow a query for the max. verbosity N and then allow levels 1-N > > Initialization > Suggestion to remove IsInitialized and IsFinalized > (not useful if we can initialize MPIT multiple times) > > Bias towards adding Fortran routines > Need more feedback from the Fortran gurus in the MPI forum > > Control variables > Very strong wish to allow the setting of control variables > In particular autotuning community is very interested and promised to > write up several concrete use cases > How to deal with setting the variables from multiple tools > a) provide lock/unlock routines > b) allow one tool to Initialize MPIT exclusively in MPIT_Init > c) Do nothing and leave races up to tools/users > How to deal with synchronizing variables > Provide additional information during query: > readonly - can never be set > sync/nosync - shows whether a variables is synchronizing or not > communicator - setting a synchronizing variables requires a global > operation on this communicator > (can be MPI_COMM_SELF for nosync variables) > Set routine takes communicator as argument > Must be called by all members of the communicator > > Extensibility > Very strong wish to have the option to extend this interface > Tools should have the ability to provide new variables > (this was brought up by several tools groups for various needs) > > Seen as essential (not just for Extensibility): Profiling routines for MPIT > > Would like to see a different iterator concept: > change from iterators to a list of variables with an integer index > (this fixes the type of the iterator) > Query the number of variables and then provide query routines to ask > for more information on each variable > Restrictions: variables can not be removed or reordered, just added > How to return query information: > Return a struct instead of a list of arguments > Cleaner ways to integrate new information in future versions > Need separate MPIT version constant to guard extensions > > Ideally: would like to give MPI an address for a variable to watch and MPI > would then integrate this variable into the MPIT variables > -> easy extension for new variables by external tools through same API > > Alternatively: tools can hide this in an PMPIT layer, but then need additional > mechanisms to allocate "dummy" handles to avoid duplicating the handle space > > > > > > > ________________________________________________________________________ > Martin Schulz, schulzm(a)llnl.gov, http://*people.llnl.gov/schulzm > CASC @ Lawrence Livermore National Laboratory, Livermore, USA > > > >

1 0

[Mpi3-tools] Next Tools WG Meeting 8/23 - MPIR & MPIT
by Martin Schulz 22 Aug '10

22 Aug '10

Hi all, Our next tools WG meeting is (as usual) on Monday 8/23 (tomorrow) at 8am PDT, 11am EDT, 5pm MSZ. Jeff created a Webex session for us, which is linked on the tools Wiki. It is: https://cisco.webex.com/ciscosales/j.php?ED=145066572&UID=0&PW=NMjcyNzAwZmU… Password: tools We have two main item for the agenda: MPIR and MPIT. MPIR: John requested some time to finish the discussion on the MPIR document before we turn the document into a standard compliant Latex document and give it for to the MPI forum forum for review (has to happen by the end of next week). MPIT: I recently gave a presentation on MPIT to the tools community and I would like to discuss the feedback from that meeting and its impact on the MPIT proposal. More detail on this are below. Talk to you tomorrow / Monday, Martin Feedback from the tools community on MPIT (based on CScADS discussions): (for those of you who were at the CScADS workshop, please feel free to add) Collection of use cases Current ideas in MPIT seem to match those use cases very well High/Low watermarks seen as quite useful Identification of exhausted resources important Naming Several name changes suggested configuration variables -> control variables (as in OpenMP) Prefixes: MPIT_CTRLVARS_... MPIT_PERFVARS_... Verbosity levels Why restrict verbosity from 1 to 5? Allow a query for the max. verbosity N and then allow levels 1-N Initialization Suggestion to remove IsInitialized and IsFinalized (not useful if we can initialize MPIT multiple times) Bias towards adding Fortran routines Need more feedback from the Fortran gurus in the MPI forum Control variables Very strong wish to allow the setting of control variables In particular autotuning community is very interested and promised to write up several concrete use cases How to deal with setting the variables from multiple tools a) provide lock/unlock routines b) allow one tool to Initialize MPIT exclusively in MPIT_Init c) Do nothing and leave races up to tools/users How to deal with synchronizing variables Provide additional information during query: readonly - can never be set sync/nosync - shows whether a variables is synchronizing or not communicator - setting a synchronizing variables requires a global operation on this communicator (can be MPI_COMM_SELF for nosync variables) Set routine takes communicator as argument Must be called by all members of the communicator Extensibility Very strong wish to have the option to extend this interface Tools should have the ability to provide new variables (this was brought up by several tools groups for various needs) Seen as essential (not just for Extensibility): Profiling routines for MPIT Would like to see a different iterator concept: change from iterators to a list of variables with an integer index (this fixes the type of the iterator) Query the number of variables and then provide query routines to ask for more information on each variable Restrictions: variables can not be removed or reordered, just added How to return query information: Return a struct instead of a list of arguments Cleaner ways to integrate new information in future versions Need separate MPIT version constant to guard extensions Ideally: would like to give MPI an address for a variable to watch and MPI would then integrate this variable into the MPIT variables -> easy extension for new variables by external tools through same API Alternatively: tools can hide this in an PMPIT layer, but then need additional mechanisms to allocate "dummy" handles to avoid duplicating the handle space ________________________________________________________________________ Martin Schulz, schulzm(a)llnl.gov, http://people.llnl.gov/schulzm CASC @ Lawrence Livermore National Laboratory, Livermore, USA

1 0

Re: [Mpi3-tools] Next MPI-3 Tools WG meeting Monday 8/26 + Schedule fornext meetings
by Martin Schulz 15 Aug '10

15 Aug '10

Hi all, No problem - let's add the MPIR discussion back on to the agenda on the 23rd, as John suggested. In the meantime, please send any feedback to the list. Time permitting, we'll also go over some issues for the MPIT interface. I presented the current API to a group of tool developers at a workshop two weeks ago and got some detailed feedback. I'll summarize this in an email later this week and can use this to guide the discussions during the WG meeting. Martin On 8/9/2010 8:47 AM, Jeff Squyres wrote: > I waited for 15 mins but no one else showed up. > > Sorry folks! > > > On Aug 9, 2010, at 11:32 AM, Jeff Squyres wrote: > > >> Sorry -- this meeting was somehow not on my calendar and I just returned from vacation today and am playing catchup. >> >> I'll join right now to see if anyone else is on... >> >> >> On Aug 9, 2010, at 11:12 AM, Mohror, Kathryn wrote: >> >> >>> Hi Bronis, >>> >>> Martin told me he asked Jeff to lead this meeting. >>> >>> Kathryn >>> >>> ------------ >>> Kathryn Mohror >>> Lawrence Livermore National Lab >>> (925)423-2997 >>> kathryn(a)llnl.gov >>> >>> >>>> -----Original Message----- >>>> From: mpi3-tools-bounces(a)lists.mpi-forum.org [mailto:mpi3-tools- >>>> bounces(a)lists.mpi-forum.org] On Behalf Of Bronis R. de Supinski >>>> Sent: Monday, August 09, 2010 8:11 AM >>>> To: mpi3-tools(a)lists.mpi-forum.org >>>> Subject: Re: [Mpi3-tools] Next MPI-3 Tools WG meeting Monday 8/26 + >>>> Schedule fornext meetings >>>> >>>> >>>> The WebEx Client is not starting correctly for me. >>>> Martin is on vacation so unless he made arrangements >>>> for someone else to run the call, I would say no... >>>> >>>> On Mon, 9 Aug 2010, John DelSignore wrote: >>>> >>>> >>>>> Hi, >>>>> >>>>> I'm the only one on the call. Is it still going to happen? >>>>> >>>>> Cheers, John D. >>>>> >>>>> >>>>> Jeff Squyres wrote: >>>>> >>>>>> On Jul 25, 2010, at 5:23 PM, Martin Schulz wrote: >>>>>> >>>>>> >>>>>>> Here is the proposed schedule for the upcoming meetings after >>>>>>> >>>> tomorrow: >>>> >>>>>>> 8/9 - Final discussion of the MPIR document >>>>>>> 8/23 - Discussion of the completed/integrated MPIT document >>>>>>> 9/6 - Labor day - no meeting >>>>>>> 9/16-9/18 - MPI forum meeting in Stuttgart >>>>>>> First reading of the MPIR document >>>>>>> 9/20 - Feedback from the MPI forum, MPIT discussions >>>>>>> >>>>>> Webex links for these meetings are now up on the wiki. >>>>>> >>>>>> >>>>> _______________________________________________ >>>>> Mpi3-tools mailing list >>>>> Mpi3-tools(a)lists.mpi-forum.org >>>>> http://***lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools >>>>> >>>>> >>>>> >>>> _______________________________________________ >>>> Mpi3-tools mailing list >>>> Mpi3-tools(a)lists.mpi-forum.org >>>> http://**lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools >>>> >>> >>> _______________________________________________ >>> Mpi3-tools mailing list >>> Mpi3-tools(a)lists.mpi-forum.org >>> http://*lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools >>> >>> >> >> -- >> Jeff Squyres >> jsquyres(a)cisco.com >> For corporate legal information go to: >> http://*www.*cisco.com/web/about/doing_business/legal/cri/ >> >> >> _______________________________________________ >> Mpi3-tools mailing list >> Mpi3-tools(a)lists.mpi-forum.org >> http://*lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-tools >> > >

1 0