<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
I have 4 gres gpus called foolsgold that I am trying to allocate, 1-to-a-job. But allocating 1 gpu allocates all gpus to that job, it seems. My batch script is:</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
#!/bin/bash
<div>#SBATCH --partition=scavenge</div>
<div>#SBATCH --qos=scavenge</div>
<div>#SBATCH --account=borrowed</div>
<div>#SBATCH --nodes=1</div>
<div>#SBATCH --tasks=1</div>
<div>#SBATCH --time=00:05:20</div>
<div>#SBATCH --gpus=foolsgold:1</div>
<div>date</div>
<div>hostname -s</div>
<div>for ((i=1;i<=1000000000;i++)) ; do a=$((i++)) ; done</div>
<div>date</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="margin:0px;font-size:12pt">And the partition definition is:</div>
<div style="margin:0px;font-size:12pt">PartitionName=scavtres Nodes=saga-test01,saga-test02 MaxTime=72:00:00 State=UP PriorityTier=0 PreemptMode=REQUEUE AllowQos=scavenge AllowAccounts=borrowed,gaia default=yes TRESBillingWeights="CPU=1.0,Mem=0.25G,GRES/foolsgold=200.0"
OverSubscribe=FORCE<br>
<br>
</div>
I have 2 compute nodes in this test cluster, each one with 4 gpus defined:</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
</div>
NodeName=saga-test01 CPUS=2 SocketsPerBoard=1 CoresPerSocket=2 ThreadsPerCore=1 RealMemory=1800 State=UNKNOWN Gres=gpu:foolsgold:4
<div> NodeName=saga-test02 CPUS=2 SocketsPerBoard=1 CoresPerSocket=2 ThreadsPerCore=1 RealMemory=1800 State=UNKNOWN Gres=gpu:foolsgold:4</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
The /etc/slurm/gres.conf on the two compute nodes:</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<div>Name=gpu Type=foolsgold File=/tmp/fg0</div>
<div>Name=gpu Type=foolsgold File=/tmp/fg1</div>
<div>Name=gpu Type=foolsgold File=/tmp/fg2</div>
<div>Name=gpu Type=foolsgold File=/tmp/fg3</div>
<br>
How can I get one gpu allocated per job?</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Thanks,</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Erik</div>
</body>
</html>