> On Feb 5, 2018, at 4:11 PM, Kilian Cavalotti > <kilian.cavalotti.w...@gmail.com> wrote: > > Hi Ryan, > > On Mon, Feb 5, 2018 at 8:06 AM, Ryan Novosielski <novos...@rutgers.edu> wrote: >> We currently use SLURM 16.05.10 and one of our staff asked how they >> can check for allocated GPUs, as you might check allocated CPUs by >> doing scontrol show node. I could have sworn that you can see both, >> but I see that only CPUs is visible. One of our staff recommended >> using sacct to see it. Is there a better way? > > scontrol -d show node <nodename> > > will display a GresUsed line, that shows how many of those Gres are > allocated, if that's what you're looking for. You'll get a count, but > not individual ids, AFAIK.
Exactly what we were looking for, thank you.
PS: Anyone know what GresDrain is? Sounds like a feature we’ve been looking for
but believed not to exist (offlining of some number of GRES). I’m assuming it’s
there but not implemented?
[novosirj@perceval1 install-files]$ scontrol -d show node cuda001
NodeName=cuda001 Arch=x86_64 CoresPerSocket=12
CPUAlloc=24 CPUErr=0 CPUTot=24 CPULoad=24.03
AvailableFeatures=(null)
ActiveFeatures=(null)
Gres=gpu:4
GresDrain=N/A
GresUsed=gpu:4
NodeAddr=cuda001 NodeHostName=cuda001 Version=16.05
OS=Linux RealMemory=128241 AllocMem=124000 FreeMem=105469 Sockets=2 Boards=1
State=ALLOCATED ThreadsPerCore=1 TmpDisk=0 Weight=4 Owner=N/A MCS_label=N/A
BootTime=2018-01-27T23:01:42 SlurmdStartTime=2018-01-28T00:17:45
CapWatts=n/a
CurrentWatts=0 LowestJoules=0 ConsumedJoules=0
ExtSensorsJoules=n/s ExtSensorsWatts=0 ExtSensorsTemp=n/s
--
____
|| \\UTGERS, |---------------------------*O*---------------------------
||_// the State | Ryan Novosielski - novos...@rutgers.edu
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
`'
signature.asc
Description: Message signed with OpenPGP
