<html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:Helvetica;
panose-1:0 0 0 0 0 0 0 0 0 0;}
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
span.apple-converted-space
{mso-style-name:apple-converted-space;}
span.EmailStyle19
{mso-style-type:personal-reply;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style>
</head>
<body lang="EN-US" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal">HI Dan,<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Ah yes now I recall better. Let’s discuss on Monday.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Howard<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in">
<p class="MsoNormal" style="margin-left:.5in"><b><span style="font-size:12.0pt;color:black">From:
</span></b><span style="font-size:12.0pt;color:black">HOLMES Daniel <d.holmes@epcc.ed.ac.uk><br>
<b>Date: </b>Friday, January 22, 2021 at 8:56 AM<br>
<b>To: </b>"Pritchard Jr., Howard" <howardp@lanl.gov><br>
<b>Cc: </b>"schulzm@in.tum.de" <schulzm@in.tum.de>, MPI Sessions working group <mpiwg-sessions@lists.mpi-forum.org><br>
<b>Subject: </b>Re: [EXTERNAL] Better handling of dynamic mixed with sessions without world<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
<p class="MsoNormal" style="margin-left:.5in">Hi Howard, <o:p></o:p></p>
<div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">I (half-)remember the discussions you mention - they mostly focused on the spawning side, rather than the being spawned side, but there was discussion of which process set(s) would need to exist for the parents
and children to find each other. The waters were muddier because we were considering a replacement for MPI_Comm_spawn (called MPI_Exec) that did not return/output an inter-communicator but expected the user to use the process set->group->comm method to find
the processes that it just created. Thus, we thought about needing multiple standardised names - whereas for this simpler approach we only need one because it is guaranteed to have unique meaning independent of the number of times spawn is called (because
only the spawned processes see it, not the processes that executed spawn).<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">This proposal would not exclude or prevent a DPM-2.0 approach that followed up on the ideas in the old slide deck.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">There is no chance of getting this into MPI-4.0, although I do wish I’d thought of this earlier, i.e. early enough to have included it at the same time as the rest of SPM-1.0 so we should push on this for the MPI-4.1
timeframe.<o:p></o:p></p>
<div>
<div>
<div>
<div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica;color:black"><br>
Cheers,<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica;color:black">Dan.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica;color:black">—<br>
Dr Daniel Holmes PhD<br>
Architect (HPC Research)<br>
<a href="mailto:d.holmes@epcc.ed.ac.uk">d.holmes@epcc.ed.ac.uk</a><br>
Phone: +44 (0) 131 651 3465<br>
Mobile: +44 (0) 7940 524 088<br>
Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica;color:black">—<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica;color:black">The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica;color:black">—<o:p></o:p></span></p>
</div>
</div>
</div>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal" style="margin-left:.5in">On 22 Jan 2021, at 15:38, Pritchard Jr., Howard <<a href="mailto:howardp@lanl.gov">howardp@lanl.gov</a>> wrote:<o:p></o:p></p>
</div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
<div>
<div style="border:dotted #FF884D 1.5pt;padding:0in 0in 0in 0in;caret-color: rgb(0, 0, 0);font-variant-caps: normal;text-align:start;-webkit-text-stroke-width: 0px;word-spacing:0px">
<p class="MsoNormal" style="margin-left:.5in;background:#FFF2E6"><b><span style="font-size:12.0pt;font-family:"Arial",sans-serif;color:black">This email was sent to you by someone outside the University.</span></b><span style="font-size:9.0pt;font-family:Helvetica;color:black">
</span><span style="font-size:9.0pt;font-family:Helvetica"><o:p></o:p></span></p>
<div>
<p class="MsoNormal" style="margin-left:.5in;background:#FFF2E6"><span style="font-size:10.0pt;font-family:"Arial",sans-serif;color:black">You should only click on links or attachments if you are certain that the email is genuine and the content is safe.</span><span style="font-size:10.0pt;font-family:"Arial",sans-serif"><o:p></o:p></span></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">Hi Dan,<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">I think this is a good idea. I’m not sure we entirely missed it though for sessions 1.0. I vaguely recall discussions about something like this in the context of mpi_exec – but it was quite a while ago. Maybe
some problems/limitations were found with this process set approach to capturing info to support spawn functionality. We should probably double check the old slide deck.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">This would be a 4.1 proposal right?<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">Howard<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
<div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0in 0in 0in">
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><b><span style="font-size:12.0pt">From:<span class="apple-converted-space"> </span></span></b><span style="font-size:12.0pt">HOLMES Daniel <<a href="mailto:d.holmes@epcc.ed.ac.uk">d.holmes@epcc.ed.ac.uk</a>><br>
<b>Date:<span class="apple-converted-space"> </span></b>Friday, January 22, 2021 at 5:06 AM<br>
<b>To:<span class="apple-converted-space"> </span></b>"Pritchard Jr., Howard" <<a href="mailto:howardp@lanl.gov">howardp@lanl.gov</a>>, Martin Schulz <<a href="mailto:schulzm@in.tum.de">schulzm@in.tum.de</a>><br>
<b>Cc:<span class="apple-converted-space"> </span></b>MPI Sessions working group <<a href="mailto:mpiwg-sessions@lists.mpi-forum.org">mpiwg-sessions@lists.mpi-forum.org</a>><br>
<b>Subject:<span class="apple-converted-space"> </span></b>[EXTERNAL] Better handling of dynamic mixed with sessions without world</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
</div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">Hi all,<span class="apple-converted-space"> </span><o:p></o:p></p>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">I just reviewed PR 506 from Rolf. It adds predefined communicator names to Appendix A. One of those is “MPI_COMM_PARENT”. It is a correct change and only indirectly relevant because it is what set off an interesting
train of thought, which I summarise below.<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">This suggests that we missed a trick when proposing Sessions v1.0 - we should have mandated the existence and meaning for an additional process set name: "<a href="mpi://MPI_COMM_PARENT">mpi://MPI_COMM_PARENT</a>”
(note the absence of “GET” in that name to match the default string name assigned by MPI to the communicator returned by the MPI_COMM_GET_PARENT procedure).<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">This process set shall contain all the processes that would be in the communicator returned by MPI_COMM_GET_PARENT, either zero processes (if you would get MPI_COMM_NULL) or the union of the local and remote groups
(if you would get an inter-communicator).<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">With this addition, MPI processes that are spawned can regain all of the functionality of the Dynamic Model without calling MPI_INIT[_THREAD], i.e. they can create the inter-communicator that they would have got
from MPI_COMM_GET_PARENT (which requires a prior call to MPI_INIT because World Model [ED: please check]).<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"> <o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">```pseudo-code<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">Create a session<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">Create an MPI_Group, groupParent, from the process set named "<a href="mpi://MPI_COMM_PARENT">mpi://MPI_COMM_PARENT</a>”<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">Create an MPI_Group, groupWorld, from the process set named “<a href="mpi://MPI_COMM_WORLD">mpi://MPI_COMM_WORLD</a>”<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">If (groupParent if MPI_GROUP_NULL) then<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"> This process was not spawned by other MPI processes<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">Else<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"> Create groupLocal from the intersection of groupWorld and groupParent<o:p></o:p></p>
</div>
</div>
<div>
<div>
<div>
<div>
<div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica"> Create groupRemote from the subtraction of groupWorld from groupParent</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica"> Create commParent using MPI_COMM_CREATE_FROM_GROUPS with groupLocal and groupRemote</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">```</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">No usage of the World Model, but we now have (a duplicate of) the inter-communicator that would have been returned by MPI_COMM_GET_PARENT if we had been permitted
to call it.</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in">For each MPI process, there is a unique “parent” process set by this definition. Multiple components doing this pseudo-code would get duplicate communicators with their own life-cycle. This contrasts with the current
way whereby freeing or disconnecting any communicator handle provided by MPI_COMM_GET_PARENT makes all other such handles stale, which is another instance of global state in MPI baked into the standardised interface and another anathema to any attempt at isolation.<o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">Of course, there is already no reason why the parent processes could not have used the Sessions Model to call MPI_COMM_SPAWN (passing in a `comm` that was derived
from the Sessions Model), so this completes the picture of how the Dynamic Model can be bolted on successfully to either the World Model (refer to MPI-3.1 and prior) or the Sessions Model (refer to MPI-4.1, which will include this addition) or a mixture (refer
to MPI-4.0, without this addition).</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">[EDIT: if we decide that MPI_COMM_GET_PARENT can already be called without/before/after the World Model, then we should add it to Table 11.1 and write some text
about it in §11.8.2. In that scenario, the above provides an implementation route for that procedure using only the Sessions Model underneath - in a similar way to the observation that it is possible to implement MPI_INIT[_THREAD] using only the Sessions Model
underneath.]</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica"> </span><o:p></o:p></p>
</div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">Cheers,</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">Dan.</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">—<br>
Dr Daniel Holmes PhD<br>
Architect (HPC Research)<br>
<a href="mailto:d.holmes@epcc.ed.ac.uk">d.holmes@epcc.ed.ac.uk</a><br>
Phone: +44 (0) 131 651 3465<br>
Mobile: +44 (0) 7940 524 088<br>
Address: Room 2.09, Bayes Centre, 47 Potterrow, Central Area, Edinburgh, EH8 9BT</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">—</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:.5in">
<p class="MsoNormal" style="margin-left:.5in"><span style="font-size:9.0pt;font-family:Helvetica">—</span><o:p></o:p></p>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</div>
</blockquote>
</div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
</div>
</body>
</html>