November 2019 - mpiwg-large-counts - mailman3.cels.anl.gov

Re: [Mpiwg-large-counts] Monday deadline!
by Jeff Squyres (jsquyres) 25 Nov '19

25 Nov '19

Sorry guys -- I'm on vacation this week, and really haven't been following the text changes you've been working on for The Embiggening... On Nov 25, 2019, at 1:18 PM, HOLMES Daniel <d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]>> wrote: Hi Jeff, Short answer: dunno! :( See my question on Slack - which issue should be checked/updated? All this has been evicted from cache a long time ago. I thought the text was there or thereabouts, but I can’t quickly locate the issue+PR+PDF combination needed for a reading announcement. Puri - can you do this today? If not, we will miss MPI-4.0, it seems. Cheers, Dan. — Dr Daniel Holmes PhD Architect (HPC Research) d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 24 Nov 2019, at 13:08, Jeff Squyres (jsquyres) via mpiwg-large-counts <mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: Puri / Tony / Dan -- Are you guys ready to send your BigCount text TOMORROW? -- Jeff Squyres jsquyres(a)cisco.com<mailto:[email protected]> _______________________________________________ mpiwg-large-counts mailing list mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts -- Jeff Squyres jsquyres(a)cisco.com<mailto:[email protected]>

1 0

Re: [Mpiwg-large-counts] Monday deadline!
by HOLMES Daniel 25 Nov '19

25 Nov '19

Hi Jeff, Short answer: dunno! :( See my question on Slack - which issue should be checked/updated? All this has been evicted from cache a long time ago. I thought the text was there or thereabouts, but I can’t quickly locate the issue+PR+PDF combination needed for a reading announcement. Puri - can you do this today? If not, we will miss MPI-4.0, it seems. Cheers, Dan. — Dr Daniel Holmes PhD Architect (HPC Research) d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 24 Nov 2019, at 13:08, Jeff Squyres (jsquyres) via mpiwg-large-counts <mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: Puri / Tony / Dan -- Are you guys ready to send your BigCount text TOMORROW? -- Jeff Squyres jsquyres(a)cisco.com<mailto:[email protected]> _______________________________________________ mpiwg-large-counts mailing list mpiwg-large-counts(a)lists.mpi-forum.org https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts

1 0

Monday deadline!
by Jeff Squyres (jsquyres) 24 Nov '19

24 Nov '19

Puri / Tony / Dan -- Are you guys ready to send your BigCount text TOMORROW? -- Jeff Squyres jsquyres(a)cisco.com

1 0

Re: [Mpiwg-large-counts] MPI Meeting Today
by Jeff Squyres (jsquyres) 13 Nov '19

13 Nov '19

I did record it; haven't gotten the link from Webex yet (sometimes it takes a while for the link to come through). On Nov 13, 2019, at 9:20 AM, Bland, Wesley via mpiwg-large-counts <mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: Since Jeff is here, I imagine he’ll take care of it, but I’ll be there too. On Nov 13, 2019, at 6:35 AM, schulzm(a)in.tum.de<mailto:[email protected]> wrote: Hi all, As mentioned before, unfortunately I won’t be able to participate today, as I will be on a plane to Denver (for SC) already. Could you please make sure (as always) that the meeting is recorded? I did send a reminder, though, to drum up participation. Thanks! Martin — Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems Department of Informatics, TU-Munich, Boltzmannstraße 3, D-85748 Garching Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) Email: schulzm(a)in.tum.de<mailto:[email protected]> <Email-signature-BSA_SC2019-450width.png> _______________________________________________ mpiwg-large-counts mailing list mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts -- Jeff Squyres jsquyres(a)cisco.com<mailto:[email protected]>

1 0

Re: [Mpiwg-large-counts] MPI Meeting Today
by Bland, Wesley 13 Nov '19

13 Nov '19

Since Jeff is here, I imagine he’ll take care of it, but I’ll be there too. On Nov 13, 2019, at 6:35 AM, schulzm(a)in.tum.de<mailto:[email protected]> wrote: Hi all, As mentioned before, unfortunately I won’t be able to participate today, as I will be on a plane to Denver (for SC) already. Could you please make sure (as always) that the meeting is recorded? I did send a reminder, though, to drum up participation. Thanks! Martin — Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems Department of Informatics, TU-Munich, Boltzmannstraße 3, D-85748 Garching Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) Email: schulzm(a)in.tum.de<mailto:[email protected]> <Email-signature-BSA_SC2019-450width.png>

1 0

MPI Meeting Today
by schulzm＠in.tum.de 13 Nov '19

13 Nov '19

Hi all, As mentioned before, unfortunately I won’t be able to participate today, as I will be on a plane to Denver (for SC) already. Could you please make sure (as always) that the meeting is recorded? I did send a reminder, though, to drum up participation. Thanks! Martin — Prof. Dr. Martin Schulz, Chair of Computer Architecture and Parallel Systems Department of Informatics, TU-Munich, Boltzmannstraße 3, D-85748 Garching Member of the Board of Directors at the Leibniz Supercomputing Centre (LRZ) Email: schulzm(a)in.tum.de

1 0

Re: [Mpiwg-large-counts] Pythonizing slides to review for Wednesday Forum Virtual Meeting
by Jim Dinan 12 Nov '19

12 Nov '19

This is really cool! +1 :) On Sun, Nov 10, 2019 at 5:27 PM Jeff Squyres (jsquyres) via mpiwg-large-counts <mpiwg-large-counts(a)lists.mpi-forum.org> wrote: > WG -- > > Remember that we have the Pythonizing + Big^H^H^HExaCount Forum Virtual > meeting this Wednesday (13 Nov). > > Here's the slides that I intend to present for the Pythonization effort: > > > https://docs.google.com/presentation/d/1dcyTuX4My-M9lwL9g2zEqAKuFQuL3dulCfN… > > Can I get a few eyeballs to review / make suggestions before the Forum > Virtual meeting? > > Our normal WG meeting is basically immediately prior to the Virtual > meeting, which is a little unfortunately (i.e., little time to react to > feedback). > > -- > Jeff Squyres > jsquyres(a)cisco.com > > _______________________________________________ > mpiwg-large-counts mailing list > mpiwg-large-counts(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts >

1 0

Re: [Mpiwg-large-counts] Pythonizing slides to review for Wednesday Forum Virtual Meeting
by Jeff Squyres (jsquyres) 12 Nov '19

12 Nov '19

(Sorry for the late reply; it was a US holiday yesterday) Just grep for "fundef" -- that's the LaTeX macro for the MPI bindings. Find any instances of those and Pythonize them (I've pushed all the Pythonizing changes that I've made, so I don't have a "soft lock" on any of the files). FWIW, I only got partway through the topology chapter. On Nov 11, 2019, at 11:22 AM, HOLMES Daniel <d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]>> wrote: Hi Jeff, I’m hoping to get a couple of hours tomorrow to do Pythonizing work. Which chapter should I do? Cheers, Dan. — Dr Daniel Holmes PhD Architect (HPC Research) d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 11 Nov 2019, at 16:14, Martin Ruefenacht <m.a.ruefenacht(a)gmail.com<mailto:[email protected]>> wrote: Hi Jeff, On slide 20, I think it is worth pointing out once we get to the point of bothering to make things pretty it is easy, compared to 400ish bindings to change. We literally just need to figure out how we want to order and indent things and then we can do it consistently everywhere immediately. It looks good! I will review the pythonizing instructions as well. I will also get the regression testing up to scratch by the virtual meeting. Martin On Mon, Nov 11, 2019 at 2:37 PM HOLMES Daniel via mpiwg-large-counts <mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: Hi Jeff, On slide 12, the “comm” parameter function call does not have a “desc” parameter, but on slide 13 the rendered output has “communicator (handle)”. The parenthetical part is auto-added, but where does the description of “communicator” come from? Is that `kind.lower()`? Is that only done for handle types? On slide 17, “Hypentization” should probably have a letter “h” in it. That’s all folks! Cheers, Dan. — Dr Daniel Holmes PhD Architect (HPC Research) d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 10 Nov 2019, at 22:26, Jeff Squyres (jsquyres) via mpiwg-large-counts <mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: WG -- Remember that we have the Pythonizing + Big^H^H^HExaCount Forum Virtual meeting this Wednesday (13 Nov). Here's the slides that I intend to present for the Pythonization effort: https://docs.google.com/presentation/d/1dcyTuX4My-M9lwL9g2zEqAKuFQuL3dulCfN… Can I get a few eyeballs to review / make suggestions before the Forum Virtual meeting? Our normal WG meeting is basically immediately prior to the Virtual meeting, which is a little unfortunately (i.e., little time to react to feedback). -- Jeff Squyres jsquyres(a)cisco.com<mailto:[email protected]> _______________________________________________ mpiwg-large-counts mailing list mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts _______________________________________________ mpiwg-large-counts mailing list mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts -- Martin -- Jeff Squyres jsquyres(a)cisco.com<mailto:[email protected]>

1 0

Re: [Mpiwg-large-counts] Pythonizing slides to review for Wednesday Forum Virtual Meeting
by HOLMES Daniel 11 Nov '19

11 Nov '19

Hi Jeff, I’m hoping to get a couple of hours tomorrow to do Pythonizing work. Which chapter should I do? Cheers, Dan. — Dr Daniel Holmes PhD Architect (HPC Research) d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 11 Nov 2019, at 16:14, Martin Ruefenacht <m.a.ruefenacht(a)gmail.com<mailto:[email protected]>> wrote: Hi Jeff, On slide 20, I think it is worth pointing out once we get to the point of bothering to make things pretty it is easy, compared to 400ish bindings to change. We literally just need to figure out how we want to order and indent things and then we can do it consistently everywhere immediately. It looks good! I will review the pythonizing instructions as well. I will also get the regression testing up to scratch by the virtual meeting. Martin On Mon, Nov 11, 2019 at 2:37 PM HOLMES Daniel via mpiwg-large-counts <mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: Hi Jeff, On slide 12, the “comm” parameter function call does not have a “desc” parameter, but on slide 13 the rendered output has “communicator (handle)”. The parenthetical part is auto-added, but where does the description of “communicator” come from? Is that `kind.lower()`? Is that only done for handle types? On slide 17, “Hypentization” should probably have a letter “h” in it. That’s all folks! Cheers, Dan. — Dr Daniel Holmes PhD Architect (HPC Research) d.holmes(a)epcc.ed.ac.uk<mailto:[email protected]> Phone: +44 (0) 131 651 3465 Mobile: +44 (0) 7940 524 088 Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT — The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. — On 10 Nov 2019, at 22:26, Jeff Squyres (jsquyres) via mpiwg-large-counts <mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]>> wrote: WG -- Remember that we have the Pythonizing + Big^H^H^HExaCount Forum Virtual meeting this Wednesday (13 Nov). Here's the slides that I intend to present for the Pythonization effort: https://docs.google.com/presentation/d/1dcyTuX4My-M9lwL9g2zEqAKuFQuL3dulCfN… Can I get a few eyeballs to review / make suggestions before the Forum Virtual meeting? Our normal WG meeting is basically immediately prior to the Virtual meeting, which is a little unfortunately (i.e., little time to react to feedback). -- Jeff Squyres jsquyres(a)cisco.com<mailto:[email protected]> _______________________________________________ mpiwg-large-counts mailing list mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts _______________________________________________ mpiwg-large-counts mailing list mpiwg-large-counts(a)lists.mpi-forum.org<mailto:[email protected]> https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts -- Martin

1 0

Re: [Mpiwg-large-counts] Pythonizing slides to review for Wednesday Forum Virtual Meeting
by Martin Ruefenacht 11 Nov '19

11 Nov '19

Hi Jeff, On slide 20, I think it is worth pointing out once we get to the point of bothering to make things pretty it is easy, compared to 400ish bindings to change. We literally just need to figure out how we want to order and indent things and then we can do it consistently everywhere immediately. It looks good! I will review the pythonizing instructions as well. I will also get the regression testing up to scratch by the virtual meeting. Martin On Mon, Nov 11, 2019 at 2:37 PM HOLMES Daniel via mpiwg-large-counts < mpiwg-large-counts(a)lists.mpi-forum.org> wrote: > Hi Jeff, > > On slide 12, the “comm” parameter function call does not have a “desc” > parameter, but on slide 13 the rendered output has “communicator (handle)”. > The parenthetical part is auto-added, but where does the description of > “communicator” come from? Is that `kind.lower()`? Is that only done for > handle types? > > On slide 17, “Hypentization” should probably have a letter “h” in it. > > That’s all folks! > > Cheers, > Dan. > — > Dr Daniel Holmes PhD > Architect (HPC Research) > d.holmes(a)epcc.ed.ac.uk > Phone: +44 (0) 131 651 3465 > Mobile: +44 (0) 7940 524 088 > Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, > EH8 9BT > — > The University of Edinburgh is a charitable body, registered in Scotland, > with registration number SC005336. > — > > On 10 Nov 2019, at 22:26, Jeff Squyres (jsquyres) via mpiwg-large-counts < > mpiwg-large-counts(a)lists.mpi-forum.org> wrote: > > WG -- > > Remember that we have the Pythonizing + Big^H^H^HExaCount Forum Virtual > meeting this Wednesday (13 Nov). > > Here's the slides that I intend to present for the Pythonization effort: > > > https://docs.google.com/presentation/d/1dcyTuX4My-M9lwL9g2zEqAKuFQuL3dulCfN… > > Can I get a few eyeballs to review / make suggestions before the Forum > Virtual meeting? > > Our normal WG meeting is basically immediately prior to the Virtual > meeting, which is a little unfortunately (i.e., little time to react to > feedback). > > -- > Jeff Squyres > jsquyres(a)cisco.com > > _______________________________________________ > mpiwg-large-counts mailing list > mpiwg-large-counts(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts > > > _______________________________________________ > mpiwg-large-counts mailing list > mpiwg-large-counts(a)lists.mpi-forum.org > https://lists.mpi-forum.org/mailman/listinfo/mpiwg-large-counts > -- Martin

1 0