<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body dir="auto">
Hi Nathan,
<div><br>
</div>
<div>It’ll be good to understand in the next call why Open MPI is losing performance with any_op compared with same_op_no_op. In MPICH, they would both perform equally mainly because of three reasons:</div>
<div><br>
</div>
<div>1. The process calling “no op” doesn’t know what the “same op” is, so it has to assume that it could be some op that is not supported in hardware, and hence rely on active messages.</div>
<div><br>
</div>
<div>2. There’s no compatibility between shared memory atomics and network atomics.</div>
<div><br>
</div>
<div>3. Derived datatypes might go through active messages (for pack/unpack reasons) even if the op is known. So even if I have contiguous datatypes I might have to assume that other processes could use derived datatypes. Issuing each contiguous segment independently
is losing too much performance.</div>
<div><br>
</div>
<div>With the proposed hints for MPI-4, we can certainly improve all these cases, but without any hint the default case should be just as good (or just as bad) for MPI-3 and MPI-4.</div>
<div><br>
</div>
<div> — Pavan<br>
<br>
<div id="AppleMailSignature" dir="ltr">Sent from my iPhone</div>
<div dir="ltr"><br>
On Nov 3, 2018, at 10:56 AM, Nathan Hjelm <<a href="mailto:hjelmn@me.com">hjelmn@me.com</a>> wrote:<br>
<br>
</div>
<blockquote type="cite">
<div dir="ltr">
<div dir="ltr"></div>
<div dir="ltr">I plan to implement the new keys in Open MPI for the next release (not v4.0.x). Just keep in mind any_op will slow accumulate, fetch and op, etc down a lot in some cases as the implemention may be unable to use network atomics. Pavan's proposal
adds another info key where you can list the atomics that will be used to hopefully reduce the number of cases where things slow down.</div>
<div dir="ltr"><br>
</div>
<div dir="ltr">-Nathan</div>
<div dir="ltr"><br>
On Nov 2, 2018, at 11:02 PM, Fujita, Hajime via mpiwg-rma <<a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a>> wrote:<br>
<br>
</div>
<blockquote type="cite">
<div dir="ltr">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@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;}
@font-face
{font-family:"Yu Mincho";
panose-1:2 2 4 0 0 0 0 0 0 0;}
@font-face
{font-family:"\@Yu Mincho";}
/* 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;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
p.msonormal0, li.msonormal0, div.msonormal0
{mso-style-name:msonormal;
mso-margin-top-alt:auto;
margin-right:0in;
mso-margin-bottom-alt:auto;
margin-left:0in;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
span.EmailStyle18
{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>
<div class="WordSection1">
<p class="MsoNormal">Got it. Thanks!<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"><b><span style="font-size:12.0pt;color:black">From: </span></b><span style="font-size:12.0pt;color:black">"Balaji, Pavan" <<a href="mailto:balaji@anl.gov">balaji@anl.gov</a>><br>
<b>Date: </b>Friday, November 2, 2018 at 3:03 PM<br>
<b>To: </b>"Fujita, Hajime" <<a href="mailto:hajime.fujita@intel.com">hajime.fujita@intel.com</a>><br>
<b>Cc: </b>"Balaji, Pavan" <<a href="mailto:balaji@anl.gov">balaji@anl.gov</a>>, MPI WG Remote Memory Access working group <<a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a>><br>
<b>Subject: </b>Re: [mpiwg-rma] Default value of accumulate_ops info key<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><br>
The invite went out to the entire WG. I've attached it again as an ics file with this email.<br>
<br>
RMA WG recurring telecon<br>
Scheduled: Nov 5, 2018 at 2:00 PM to 3:00 PM<br>
To join the Meeting:<br>
<a href="https://bluejeans.com/173205081/3141592">https://bluejeans.com/173205081/3141592</a><br>
<br>
To join via Room System:<br>
Video Conferencing System: bjn.vc -or-199.48.152.152<br>
Meeting ID : 173205081<br>
Participant Passcode : 3141592<br>
<br>
To join via phone :<br>
1) Dial:<br>
+1.408.317.9254 (US (San Jose))<br>
+1.866.226.4650 (US Toll Free)<br>
(see all numbers - <a href="http://bluejeans.com/numbers">http://bluejeans.com/numbers</a>)<br>
2) Enter Conference ID : 173205081<br>
3) Enter Participant Passcode : 3141592<o:p></o:p></p>
</div>
</div>
<div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><br>
<br>
-- Pavan<br>
<br>
> On Nov 2, 2018, at 2:37 PM, Fujita, Hajime <<a href="mailto:hajime.fujita@intel.com">hajime.fujita@intel.com</a>> wrote:<br>
> <br>
> Hi Pavan,<br>
> <br>
> So the next meeting is at 11/5 (Mon) 2pm central? I don't see a calendar invite.<br>
> <br>
> Thanks,<br>
> Hajime<br>
> <br>
> On 11/2/18, 12:11 PM, "mpiwg-rma on behalf of Balaji, Pavan via mpiwg-rma" <<a href="mailto:mpiwg-rma-bounces@lists.mpi-forum.org">mpiwg-rma-bounces@lists.mpi-forum.org</a> on behalf of
<a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a>> wrote:<br>
> <br>
> Hi Joseph,<br>
> <br>
> I don't know about MPI-3.2, but if the Forum accepts the proposal, I believe we have started pushing out draft standards which can be used. Having said that, I don't know if MPI implementations will implement draft standards or wait till the final standard
is out.<br>
> <br>
> The last telecon was canceled. We have one more this coming Monday.<br>
> <br>
> -- Pavan<br>
> <br>
>> On Nov 2, 2018, at 10:08 AM, Joseph Schuchart via mpiwg-rma <<a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a>> wrote:<br>
>> <br>
>> All,<br>
>> <br>
>> While working on a code that makes use of MPI accumulate operations, I stumbled upon reading the description of the `accumulate_ops` info key. The standard (v3.1) states "The default is same_op_no_op." My interpretation is that this is the value used if
the info key is not specified (based on my understanding of the term `default`). How would I specify my intent to use any accumulate operation? From looking at one implementation, the default is to expect the use of any operation but that seems inconsistent
and confusing for users.<br>
>> <br>
>> I remember the discussion of a proposal for v4.0 to add `any_op` in a recent phone conference but I couldn't find that proposal anywhere online. Would it be possible to fix this issue in v3.2 by removing the sentence quoted above?<br>
>> <br>
>> On another note: Based on the invitation sent out by Pavan Balaji I tried to join the biweekly phone conference on Monday last week (10/22) at 2pm central time but the conference room was empty. Is this still the current schedule? Maybe I just got the timezones
messed up?<br>
>> <br>
>> Many thanks in advance,<br>
>> Joseph Schuchart<br>
>> -- <br>
>> Dipl.-Inf. Joseph Schuchart<br>
>> High Performance Computing Center Stuttgart (HLRS)<br>
>> Nobelstr. 19<br>
>> D-70569 Stuttgart<br>
>> <br>
>> Tel.: +49(0)711-68565890<br>
>> Fax: +49(0)711-6856832<br>
>> E-Mail: <a href="mailto:schuchart@hlrs.de">schuchart@hlrs.de</a><br>
>> _______________________________________________<br>
>> mpiwg-rma mailing list<br>
>> <a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a><br>
>> <a href="https://lists.mpi-forum.org/mailman/listinfo/mpiwg-rma">https://lists.mpi-forum.org/mailman/listinfo/mpiwg-rma</a><br>
> <br>
> _______________________________________________<br>
> mpiwg-rma mailing list<br>
> <a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a><br>
> <a href="https://lists.mpi-forum.org/mailman/listinfo/mpiwg-rma">https://lists.mpi-forum.org/mailman/listinfo/mpiwg-rma</a><br>
> <br>
> <o:p></o:p></p>
</div>
</div>
</div>
</div>
</blockquote>
<blockquote type="cite">
<div dir="ltr"><span>_______________________________________________</span><br>
<span>mpiwg-rma mailing list</span><br>
<span><a href="mailto:mpiwg-rma@lists.mpi-forum.org">mpiwg-rma@lists.mpi-forum.org</a></span><br>
<span><a href="https://lists.mpi-forum.org/mailman/listinfo/mpiwg-rma">https://lists.mpi-forum.org/mailman/listinfo/mpiwg-rma</a></span><br>
</div>
</blockquote>
</div>
</blockquote>
</div>
</body>
</html>