[HTCondor-users] how to use rank with memory

[Myunggi Yi <myunggi@xxxxxxxxxxxxx>]

Dear users,

I have installed HTCondor 9.0.1

I want the jobs to run on more memory machines.

I submitted a job with the following script, but the job always goes to node07, which has less memory.

How can I achieve my goal?

Thank you for your help.

==========================

Executable      Â= train.sh
Log          = $outfile.log
Error         = $outfile.err
Output        Â= $outfile.out
# NFS
+IwdFlusNFSCache   Â= False
Should_transfer_files = no
GetEnv        Â= True
# for ML jobs
Request_GPUs     Â= 1
Requirements     Â= CUDACapability > 3.0
Rank         Â= memory
# Prevent re-run
periodic_remove    = JobStatus == 1 && NumJobStarts > 0
# Email
Notification     Â= Always
Notify_user      = myunggi@xxxxxxxxxxxxx
Queue

=================================

The following is my condor_status

node01.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 80222 Â0+22:49:4
node02.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 64094 Â0+22:49:4
node03.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 64094 Â0+22:49:4
node04.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 64094 Â0+19:18:0
node05.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 64094 Â0+22:49:4
node07.synapse   LINUX   ÂX86_64 Claimed  Busy   Â0.280 31892 Â0+00:00:3
node08.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 31892 Â0+22:49:3
node09.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 31892 Â0+22:49:3
node10.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 31892 Â0+22:49:2
node11.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 15779 Â0+15:34:3
node12.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 64042 Â0+22:49:3
node13.synapse   LINUX   ÂX86_64 Unclaimed Idle   Â0.000 15819 Â0+22:44:3

Best regards,