<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;}
@font-face
{font-family:"Apple Color Emoji";
panose-1:0 0 0 0 0 0 0 0 0 0;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
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:612.0pt 792.0pt;
margin:70.85pt 70.85pt 70.85pt 70.85pt;}
div.WordSection1
{page:WordSection1;}
--></style>
</head>
<body lang="FR" link="blue" vlink="purple" style="word-wrap:break-word">
<div class="WordSection1">
<p class="MsoNormal"><span lang="EN-US" style="mso-fareast-language:EN-US">Sure.<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="mso-fareast-language:EN-US">Servers with both IO-bound and CPU-bound requests are what I see in the real world.
</span><span lang="EN-US" style="font-family:"Apple Color Emoji";mso-fareast-language:EN-US">😊</span><span lang="EN-US" style="mso-fareast-language:EN-US"><o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="mso-fareast-language:EN-US"><o:p> </o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="mso-fareast-language:EN-US">Thanks<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="mso-fareast-language:EN-US">Arnaud<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="mso-fareast-language:EN-US"><o:p> </o:p></span></p>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">Over the years we’ve been working on virtual threads, we’ve made lots of simulations.
</span>But the features we have are not features that address artificial simulations, but only those of them that correspond to real problems our real users face in the real world. What we need to improve things aren’t more simulations, but reports from real
systems (or simulations that try to mimic behaviour observed in a real system). <o:p>
</o:p></p>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt">— Ron <o:p></o:p></p>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><br>
<br>
<o:p></o:p></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal" style="margin-left:35.4pt">On 6 Jan 2023, at 19:45, Arnaud Masson <<a href="mailto:arnaud.masson@fr.ibm.com">arnaud.masson@fr.ibm.com</a>> wrote:<o:p></o:p></p>
</div>
<p class="MsoNormal" style="margin-left:35.4pt"><o:p> </o:p></p>
<div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">I can’t see how stop-the-world effect can be avoided once all your carriers are busy with non-switchable CPU-bound tasks. Maybe I’m missing something<span class="apple-converted-space"> </span></span><span lang="EN-US" style="font-family:"Apple Color Emoji"">😊</span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">Not very different from other pinning problems (JNI...), except the argument that 100% CPU usage should never occur so not a problem.</span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">I will try to make some test to simulate and post the result here.</span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">Thanks</span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">Arnaud</span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">I don’t think that increasing the scheduler’s parallelism would help, nor do I think you’d see a “stop-the-world”, but again, these hypotheses are just not actionable. There’s nothing we can
do to address them. When you find a problem, please report it and we’ll investigate what can be done.</span><o:p></o:p></p>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"><br>
<br>
<br>
</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">— Ron</span><o:p></o:p></p>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"><br>
<br>
<br>
</span><o:p></o:p></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">On 6 Jan 2023, at 19:11, Arnaud Masson <</span><a href="mailto:arnaud.masson@fr.ibm.com"><span lang="EN-US">arnaud.masson@fr.ibm.com</span></a><span lang="EN-US">> wrote:</span><o:p></o:p></p>
</div>
</div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
<div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">I don’t think having 100% CPU usage on a pod is enough to justify a “stop-the-world” effect on Loom scheduling for the other tasks.</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">Also 100% is the extreme case, but there can be 75% CPU usage, meaning only 1 carrier left for all other tasks in my example.</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">Again not a blocker I guess, just have to increase the carrier count to mitigate, but it’s good old native thread sizing again where it should not be really needed.</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">“Time-sharing would make those expensive tasks complete in a lot more than 10 seconds”:</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">I understand there would be switching overhead (so it’s slower), but I don’t understand why it would be<span class="apple-converted-space"> </span><i>much</i><span class="apple-converted-space"> </span>slower
if there are few of them like in my example.</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">thanks</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US">Arnaud</span><o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"><span lang="EN-US"> </span><o:p></o:p></p>
</div>
</div>
</div>
</blockquote>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"> <o:p></o:p></p>
</div>
</div>
<div>
<div style="margin-left:35.4pt">
<p class="MsoNormal" style="margin-left:35.4pt"> <o:p></o:p></p>
</div>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-left:35.4pt"><span style="font-size:9.0pt;font-family:Helvetica">Unless otherwise stated above:<br>
<br>
Compagnie IBM France<br>
Siège Social : 17, avenue de l'Europe, 92275 Bois-Colombes Cedex<br>
RCS Nanterre 552 118 465<br>
Forme Sociale : S.A.S.<br>
Capital Social : 664 069 390,60 €<br>
SIRET : 552 118 465 03644 - Code NAF 6203Z<o:p></o:p></span></p>
</div>
</div>
</blockquote>
</div>
<p class="MsoNormal" style="margin-left:35.4pt"><o:p> </o:p></p>
</div>
</div>
<DIV>
Unless otherwise stated above:<BR>
<BR>
Compagnie IBM France<BR>
Siège Social : 17, avenue de l'Europe, 92275 Bois-Colombes Cedex<BR>
RCS Nanterre 552 118 465<BR>
Forme Sociale : S.A.S.<BR>
Capital Social : 664 069 390,60 €<BR>
SIRET : 552 118 465 03644 - Code NAF 6203Z<BR>
</DIV></body>
</html>