<html><head><meta http-equiv="Content-Type" content="text/html charset=windows-1252"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">Rolf,<div><br></div><div>Thanks for putting together #456; it helps to have this all in one place. The first thing that we need to do is decide between options A and B, noting that the increased synchronization required of (A) applies only to shared memory windows (and this still does not require a barrier, though that’s likely to be the easiest implementation). My preference is to keep the discussion of remote load/store separate from MPI RMA operations and to provide text specific for this case. I don’t believe that there will be enough members of the RMA group at the next meeting to come to a consensus on this, and the issues are subtle, so I don’t think it will be possible to get this resolved before or at the next meeting.</div><div><br></div><div>Bill</div><div><br><div apple-content-edited="true">
<span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: -webkit-auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; "><span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: -webkit-auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; "><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; "><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div style="font-size: 12px; ">William Gropp</div><div style="font-size: 12px; ">Director, Parallel Computing Institute</div></div></span><span class="Apple-style-span" style="font-size: 12px; ">Thomas M. Siebel Chair in Computer Science</span><span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; "><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div><div style="font-size: 12px; ">University of Illinois Urbana-Champaign</div></div><div><br></div></div></span><br class="Apple-interchange-newline"></div></span><br class="Apple-interchange-newline"></span><br class="Apple-interchange-newline">
</div>
<br><div><div>On Aug 15, 2014, at 3:06 AM, Rolf Rabenseifner <<a href="mailto:rabenseifner@hlrs.de">rabenseifner@hlrs.de</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">Dear Bill and Rajeev,<br><br>the Sep. 1 deadline is in 17 days.<br>I know that the original plan from the last telecon<br>2 weeks ago was that you, Bill make an initial proposal.<br>I didn't felt that I would be able to do such a proposal.<br><br>After all the helpful discussion by email,<br>I felt that I now have enough information.<br>I hope that it was okay for you that I tried to put all<br>what I knew together into the proposal #456.<br><br>To get further progress on this problem, I expect we <br>should have another telecon in the first half of next<br>week (already less than 14 days before the deadline).<br><br>What is your opinion?<br><br>I'm thankful that I was teached about the background, <br>especially by Dave, Pavan, Bronis, Bill Long, and you, Bill.<br><br>Therefore I hope that my proposal is technically<br>a correct solution (but still incomplete for the two <br>FLUSH operations).<br>I hope that you or somebody else can fill this gap.<br><br>Best regards<br>Rolf <br><br>----- Original Message -----<br><blockquote type="cite">From: "Rajeev Thakur" <<a href="mailto:thakur@mcs.anl.gov">thakur@mcs.anl.gov</a>><br>To: "MPI WG Remote Memory Access working group" <<a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a>><br>Sent: Monday, July 28, 2014 4:26:05 PM<br>Subject: Re: [mpiwg-rma] doodle poll for RMA working group telecon<br><br>A reminder of this telecon on Wednesday. Please read ticket 437 and<br>the other tickets mentioned in it.<br><a href="https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/437">https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/437</a><br><br>Rajeev<br><br>On Jul 19, 2014, at 5:11 PM, Rajeev Thakur <thakur@mcs.anl.gov><br>wrote:<br><br><blockquote type="cite">OK, let's have the telecon on Wed, July 30, 9.00am Central Time<br>(7am Pacific, 4pm German time).<br><br>We can use this telecon number: 1-567-314-8345, passcode:<br>262144<br><br>The agenda is ticket 437 and other tickets mentioned therein.<br>https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/437<br><br>Rajeev<br><br><br>On Jul 18, 2014, at 2:26 PM, William Gropp <wgropp@illinois.edu><br>wrote:<br><br><blockquote type="cite">Starting with the first available time, 9 CDT on July 30, is fine<br>with me.<br>Bill<br><br>William Gropp<br>Director, Parallel Computing Institute<br>Thomas M. Siebel Chair in Computer Science<br>University of Illinois Urbana-Champaign<br><br><br><br><br><br>On Jul 18, 2014, at 10:45 AM, Rolf Rabenseifner wrote:<br><br><blockquote type="cite">Dear all,<br><br>Rajeev, thank you for starting the doodle.<br>I expect that you are still on vacation.<br>Therefore a short summary before the weekend starts (in Europe):<br><br>As I can see at http://doodle.com/hx3p4cbwfcyszam6<br>there are only three slots available where all can come together:<br><br>Wednesday July 30,<br>- 09:00-10:00 USA Central time<br>- 16:00-17:00 Berlin timezone<br><br>Wednesday July 30,<br>- 12:00-13:00 USA Central time<br>- 19:00-20:00 Berlin timezone<br>(this slot is also available, because my evening-date was<br>removed)<br><br>and Thursday July 31,<br>- 12:00-13:00 USA Central time<br>- 19:00-20:00 Berlin timezone<br><br><blockquote type="cite">We need to discuss a few tickets over telecon. Let us start with<br>a<br>telecon for ticket 437 that Rolf has created:<br>https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/437. It is<br>to<br>clarify whether remote accesses to shared memory are to be<br>considered similar to RMA operations or local loads/stores. A<br>few<br>related tickets are linked there.<br></blockquote><br>I would recommend that we start with the first slot and<br>we will see, whether we need to continue with the other slots.<br>Otherwise those other slots could be used for other RMA tickets.<br><br>Okay?<br><br>Participants on the doodle were:<br>- Rajeev Thakur<br>- Pavan Balaji<br>- Jeff Hammond<br>- Rolf Rabenseifner<br>- Bill Gropp<br>- Jim Dinan<br>- Hubert Ritzdorf<br><br>I hope that other members of the RMA WG can also participate,<br>especially if you will go to the Japanese MPI forum meeting.<br><br>Best regards<br>Rolf<br><br><br>----- Original Message -----<br><blockquote type="cite">From: "Rajeev Thakur" <thakur@mcs.anl.gov><br>To: "MPI WG Remote Memory Access working group"<br><mpiwg-rma@lists.mpi-forum.org><br>Sent: Monday, July 14, 2014 8:39:12 PM<br>Subject: [mpiwg-rma] doodle poll for RMA working group telecon<br><br>We need to discuss a few tickets over telecon. Let us start with<br>a<br>telecon for ticket 437 that Rolf has created:<br>https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/437. It is<br>to<br>clarify whether remote accesses to shared memory are to be<br>considered similar to RMA operations or local loads/stores. A<br>few<br>related tickets are linked there.<br><br>Please indicate your availability in the weeks of July 21 and 28<br>in<br>the doodle poll at http://doodle.com/hx3p4cbwfcyszam6. I have<br>listed<br>only a few slots on each day to accommodate European and Pacific<br>time zones.<br><br>Thanks,<br>Rajeev<br><br>_______________________________________________<br>mpiwg-rma mailing list<br>mpiwg-rma@lists.mpi-forum.org<br>http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma<br><br></blockquote><br>--<br>Dr. Rolf Rabenseifner . . . . . . . . . .. email<br>rabenseifner@hlrs.de<br>High Performance Computing Center (HLRS) . phone<br>++49(0)711/685-65530<br>University of Stuttgart . . . . . . . . .. fax ++49(0)711 /<br>685-65832<br>Head of Dpmt Parallel Computing . . .<br>www.hlrs.de/people/rabenseifner<br>Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room<br>1.307)<br>_______________________________________________<br>mpiwg-rma mailing list<br>mpiwg-rma@lists.mpi-forum.org<br>http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma<br></blockquote><br>_______________________________________________<br>mpiwg-rma mailing list<br>mpiwg-rma@lists.mpi-forum.org<br>http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma<br></blockquote><br>_______________________________________________<br>mpiwg-rma mailing list<br>mpiwg-rma@lists.mpi-forum.org<br>http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma<br></blockquote><br>_______________________________________________<br>mpiwg-rma mailing list<br>mpiwg-rma@lists.mpi-forum.org<br>http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-rma<br><br></blockquote><br>-- <br>Dr. Rolf Rabenseifner . . . . . . . . . .. email <a href="mailto:rabenseifner@hlrs.de">rabenseifner@hlrs.de</a><br>High Performance Computing Center (HLRS) . phone ++49(0)711/685-65530<br>University of Stuttgart . . . . . . . . .. fax ++49(0)711 / 685-65832<br>Head of Dpmt Parallel Computing . . . <a href="http://www.hlrs.de/people/rabenseifner">www.hlrs.de/people/rabenseifner</a><br>Nobelstr. 19, D-70550 Stuttgart, Germany . . . . (Office: Room 1.307)<br></blockquote></div><br></div></body></html>