- 16 Apr, 2009 2 commits
- 15 Apr, 2009 7 commits
-
-
Moe Jette authored
which jobs can be submitted to this partition. From Matthieu Hautreux, CEA.
-
Moe Jette authored
-
Moe Jette authored
-
Don Lipari authored
-
Moe Jette authored
-
Danny Auble authored
-
Danny Auble authored
-
- 14 Apr, 2009 12 commits
-
-
Moe Jette authored
-
Danny Auble authored
-
Danny Auble authored
-
Moe Jette authored
-
Danny Auble authored
-
Danny Auble authored
-
Moe Jette authored
-
Moe Jette authored
-
Danny Auble authored
-
Moe Jette authored
used to set SLURM_CPUS_ON_NODE env var and compute memory limit. We compare the allocated CPUs information passed in the job step credential (typically cores) and compare with the CPUs configured on the node and scale the value as needed.
-
Moe Jette authored
all job priorities are very low.
-
Danny Auble authored
-
- 11 Apr, 2009 6 commits
-
-
Moe Jette authored
-
Moe Jette authored
I'll need to modify the test for systems with >1 ThreadPerCore configured
-
Moe Jette authored
found in testing that new code: - avoid draining node if stdout/err file can't be created - correction in logic for >1 thread per core in setting a job's SLURM_CPUS_ON_NODE env var - Fix for uneven task distributions on heterogeneous clusters (e.g. 4 CPUs on the first node and 2 CPUs on the second, a task_count=2 resulted in cpus_per_task=3, and previously allocated both tasks to the first node, now does one task per node)
-
-
Danny Auble authored
-
Danny Auble authored
-
- 10 Apr, 2009 4 commits
-
-
Danny Auble authored
-
Danny Auble authored
-
Moe Jette authored
-
Moe Jette authored
-
- 09 Apr, 2009 9 commits
-
-
Moe Jette authored
(was using day of month=0 instead of 1)
-
Moe Jette authored
-
Moe Jette authored
-
Joseph P. Donaghy authored
-
Moe Jette authored
-
Danny Auble authored
-
-
Moe Jette authored
-
Moe Jette authored
Mostly a matter of moving some logic to a new module.
-