Hi,

 

I am working on a PhD on supercomputer scheduling and I am investigating if
SLURM has the capabilities I am looking to use in a scheduler I want to
develop and test.

 

The main requirements are that the scheduler interface can suspend resume
jobs, multiple jobs can be scheduled on a node (including mixing suspended
and non-suspended nodes), and that the scheduler can access up to date stats
such as memory use, CPU use etc. per job per node.

 

Can you advise me if SLURM meets these requirements?

 

Thanks,

Leith Bade

[email protected]

 

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to