<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Hi all,</p>
<p>I sent this mail from a bad email address this week-end. I
apologize if it is published duplicate (but not found in the
archive yet).<br>
</p>
<p>May be this is a basic question but I'm stuck with it. I'm quite
new in managing a small cluster with slurm instead of a local
batch scheduler. On the nodes I've set memory limits in
slurm.conf.<br>
</p>
<blockquote>
<p>DefMemPerCPU=2048<br>
MaxMemPerCPU=4096</p>
</blockquote>
<p>Requesting 1.2GB of RAM works:</p>
<blockquote>
<p>srun --ntasks-per-node=1 --mem-per-cpu=1500M -p tenibre-gpu
--pty bash -i</p>
</blockquote>
<p>and my testcase can allocate until 1.5GB:</p>
<blockquote>./a.out <br>
allocation de 1000Mo.........Ok<br>
....<br>
allocation de 1419Mo.........Ok<br>
allocation de 1524Mo.........Ok<br>
Killed<br>
</blockquote>
<p>Now I would like to use more memory than MaxMemPerCPU:</p>
<blockquote>
<p>srun --ntasks-per-node=1 --mem-per-cpu=12G -p tenibre-gpu --pty
bash -i<br>
</p>
</blockquote>
<p>So, if I understand the documentation, as mem-per-cpu >
MaxMemPerCPU this is a limitation applied to the task and it
agregates cpu and memory. The squeue command show 3 cpu agregated
on the node to reach the 3*MaxMemPerCPU memory requested so all
seams correct.<br>
</p>
<blockquote>
<p> JOBID PARTITION NAME USER ST
TIME START_TIME TIME_LIMIT CPUS NODELIST(REASON)<br>
497 tenibre-gpu bash begou R 1:23
2021-03-20T14:42:47 12:00:00 3 tenibre-gpu-0<br>
</p>
</blockquote>
<p>But my task is unable to exceed the MaxMemPerCPU value ?</p>
<blockquote>
<p>./a.out <br>
allocation de 1000Mo.........Ok<br>
....<br>
</p>
allocation de 4145Mo.........Ok<br>
allocation de 4250Mo.........Ok<br>
Killed<br>
</blockquote>
<p>So, I'm wrong somewhere but ? </p>
<p>Running the testcase in a ssh sessions (ssh as root then su as a
basic user) allows using more memory so it is related to my bad
slurm setup/use<br>
</p>
<p>Patrick</p>
</body>
</html>