<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=iso-8859-2">
<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;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
span.DefaultFontHxMailStyle
{mso-style-name:"Default Font HxMail Style";
font-family:"Calibri",sans-serif;
color:windowtext;
font-weight:normal;
font-style:normal;
text-decoration:none none;}
.MsoChpDefault
{mso-style-type:export-only;}
@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="#954F72" style="word-wrap:break-word">
<div class="WordSection1">
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt">It should be possible,
<o:p></o:p></span></span></p>
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt">If the test is supersmall, it can be a little bit difficult though. <o:p></o:p></span></span></p>
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt">For small test with not a big number of plane waves and bands, one needs to use as much pools as possible.
<o:p></o:p></span></span></p>
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt">But program need to run with 1 MPI rank per GPU, it is possible to superscribe the GPUs, but you need to have nvidia MPS on the node for this to be efficient.
<o:p></o:p></span></span></p>
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt"><o:p> </o:p></span></span></p>
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt"><o:p> </o:p></span></span></p>
<p class="MsoNormal">Sent from <a href="https://go.microsoft.com/fwlink/?LinkId=550986">
Mail</a> for Windows</p>
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt"><o:p> </o:p></span></span></p>
<div style="mso-element:para-border-div;border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in">
<p class="MsoNormal" style="border:none;padding:0in"><b>From: </b><a href="mailto:malwi45@gmail.com">Małgorzata Wierzbowska</a><br>
<b>Sent: </b>Wednesday, June 21, 2023 10:00 AM<br>
<b>To: </b><a href="mailto:developers@lists.quantum-espresso.org">developers@lists.quantum-espresso.org</a><br>
<b>Subject: </b>[QE-developers] QE GPU test results</p>
</div>
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt"><o:p> </o:p></span></span></p>
<div>
<div>
<p class="MsoNormal"> <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"> Dear QE Team,<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">we tested v.7.2 on Athena computer with GPU<br>
<a href="https://www.cyfronet.pl/en/19073,artykul,athena.html">https://www.cyfronet.pl/en/19073,artykul,athena.html</a><br>
<br>
The simple case for nscf with large kmesh that runs 37 min at 96 cpu on Ares<br>
<a href="https://www.cyfronet.pl/en/computers/18827,artykul,ares_supercomputer.html">https://www.cyfronet.pl/en/computers/18827,artykul,ares_supercomputer.html</a><br>
<br>
gives the following results on Athena:<br>
<br>
1-1-1-1: PWSCF : 1h 6m CPU 1h13m WALL<br>
<br>
1-2-1-2: PWSCF : 1h 3m CPU 1h 8m WALL<br>
<br>
1-4-1-4: PWSCF : 1h 1m CPU 1h 6m WALL<br>
<br>
1-8-1-8: PWSCF : 1h12m CPU 1h17m WALL<br>
<br>
Where the configurations 1-2-3-4 mean: <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">1 node 2 tasks (processes MPI), 3 cpus-per-task, 4 cards GPGPU<br>
<br>
Is it possible to get it better?<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">With best regards,<br>
Malgorzata<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<div>
<div>
<p class="MsoNormal"><span style="color:#999999;background:white">dr hab. Malgorzata Wierzbowska, prof. IHPP PAS</span><o:p></o:p></p>
<div>
<p style="margin-bottom:0in"><span style="color:#999999;background:white">Institute of High Pressure Physics (Unipress)</span></p>
<p style="margin-bottom:0in"><span style="color:#999999;background:white">Polish Academy of Sciences
</span></p>
<p style="margin-bottom:0in"><span style="color:#999999;background:white">Sokolowska 29/37, 01-142 Warsaw, Poland</span></p>
</div>
</div>
</div>
</div>
</div>
<p style="margin-bottom:0in"><span style="color:#999999;background:white">email: <a href="mailto:malwi@gmail.com" target="_blank">
malwi45@gmail.com</a> </span></p>
<p class="MsoNormal"><span class="DefaultFontHxMailStyle"><span style="font-size:12.0pt"><o:p> </o:p></span></span></p>
</div>
</body>
</html>