Resource Queues¶

In order to use all the nodes efficiently and in a fair share fashion, resources are distributed in queues (partitions). Queues group nodes into logical sets, each of which has an assortment of constraints such as job size limit, job time limit, users permitted to use it, etc.

To determine what partitions exist on the system, what nodes they include, and general system state. use the sinfo command.

sinfo -s

sinfo man page

ARIS queue (partition) overview:

Queue Table¶

PARTITION	DESCRIPTION	AVAIL	TIMELIMIT	NODES	NODELIST
compute	Compute nodes w/o accelerator	up	2-00:00:00	48	m[01-48]
fat	Fat compute nodes	up	2-00:00:00	16	f[01-16]
gpu	GPU accelerated nodes	up	2-00:00:00	2	a[02-03]
mig	MIG GPU accelerated node	up	2-00:00:00	1	a01

compute queue: Is intended to run parallel jobs on the THIN compute nodes.
fat queue: Is dedicated to run parallel jobs on the FAT compute nodes.

gpu queue: Provides access on the GPU accelerated nodes.

The scontrol command can be used to report more detailed information partitions and configuration

:$ scontrol show partition

PartitionName=compute
   AllowGroups=ALL AllowAccounts=ALL AllowQos=ALL
   AllocNodes=ALL Default=YES QoS=N/A
   DefaultTime=NONE DisableRootJobs=NO ExclusiveUser=NO ExclusiveTopo=NO GraceTime=0 Hidden=NO
   MaxNodes=UNLIMITED MaxTime=2-00:00:00 MinNodes=0 LLN=NO MaxCPUsPerNode=UNLIMITED MaxCPUsPerSocket=UNLIMITED
   Nodes=m[01-48]
   PriorityJobFactor=1 PriorityTier=1 RootOnly=NO ReqResv=NO OverSubscribe=NO
   OverTimeLimit=NONE PreemptMode=OFF
   State=UP TotalCPUs=6144 TotalNodes=48 SelectTypeParameters=NONE
   JobDefaults=(null)
   DefMemPerNode=UNLIMITED MaxMemPerNode=507904
   TRES=cpu=6144,mem=23808G,node=48,billing=6144

PartitionName=fat
   AllowGroups=ALL AllowAccounts=ALL AllowQos=ALL
   AllocNodes=ALL Default=NO QoS=N/A
   DefaultTime=NONE DisableRootJobs=NO ExclusiveUser=NO ExclusiveTopo=NO GraceTime=0 Hidden=NO
   MaxNodes=UNLIMITED MaxTime=2-00:00:00 MinNodes=0 LLN=NO MaxCPUsPerNode=UNLIMITED MaxCPUsPerSocket=UNLIMITED
   Nodes=f[01-16]
   PriorityJobFactor=1 PriorityTier=1 RootOnly=NO ReqResv=NO OverSubscribe=NO
   OverTimeLimit=NONE PreemptMode=OFF
   State=UP TotalCPUs=2048 TotalNodes=16 SelectTypeParameters=NONE
   JobDefaults=(null)
   DefMemPerNode=UNLIMITED MaxMemPerNode=1015808
   TRES=cpu=2048,mem=15.50T,node=16,billing=2048

PartitionName=gpu
   AllowGroups=ALL AllowAccounts=ALL AllowQos=ALL
   AllocNodes=ALL Default=NO QoS=N/A
   DefaultTime=NONE DisableRootJobs=NO ExclusiveUser=NO ExclusiveTopo=NO GraceTime=0 Hidden=NO
   MaxNodes=UNLIMITED MaxTime=2-00:00:00 MinNodes=0 LLN=NO MaxCPUsPerNode=UNLIMITED MaxCPUsPerSocket=UNLIMITED
   Nodes=a[02-03]
   PriorityJobFactor=1 PriorityTier=1 RootOnly=NO ReqResv=NO OverSubscribe=NO
   OverTimeLimit=NONE PreemptMode=OFF
   State=UP TotalCPUs=256 TotalNodes=2 SelectTypeParameters=NONE
   JobDefaults=(null)
   DefMemPerNode=UNLIMITED MaxMemPerNode=507904
   TRES=cpu=256,mem=992G,node=2,billing=256,gres/gpu=8,gres/gpu:a100=8

PartitionName=mig
   AllowGroups=ALL AllowAccounts=ALL AllowQos=ALL
   AllocNodes=ALL Default=NO QoS=N/A
   DefaultTime=NONE DisableRootJobs=NO ExclusiveUser=NO ExclusiveTopo=NO GraceTime=0 Hidden=NO
   MaxNodes=UNLIMITED MaxTime=2-00:00:00 MinNodes=0 LLN=NO MaxCPUsPerNode=UNLIMITED MaxCPUsPerSocket=UNLIMITED
   Nodes=a01
   PriorityJobFactor=1 PriorityTier=1 RootOnly=NO ReqResv=NO OverSubscribe=NO
   OverTimeLimit=NONE PreemptMode=OFF
   State=UP TotalCPUs=128 TotalNodes=1 SelectTypeParameters=NONE
   JobDefaults=(null)
   DefMemPerNode=UNLIMITED MaxMemPerNode=507904
   TRES=cpu=128,mem=496G,node=1,billing=128,gres/gpu=16,gres/gpu:1g.10gb=8,gres/gpu:2g.20gb=4,gres/gpu:3g.40gb=4

:$ scontrol show node m01

NodeName=m01 Arch=x86_64 CoresPerSocket=64 
   CPUAlloc=0 CPUEfctv=128 CPUTot=128 CPULoad=0.00
   AvailableFeatures=(null)
   ActiveFeatures=(null)
   Gres=(null)
   NodeAddr=m01 NodeHostName=m01 Version=24.11.5
   OS=Linux 5.14.0-503.35.1.el9_5.x86_64 #1 SMP PREEMPT_DYNAMIC Thu Apr 3 12:12:16 UTC 2025 
   RealMemory=507904 AllocMem=0 FreeMem=507277 Sockets=2 Boards=1
   State=IDLE ThreadsPerCore=1 TmpDisk=0 Weight=1 Owner=N/A MCS_label=N/A
   Partitions=compute 
   BootTime=2025-05-15T14:02:30 SlurmdStartTime=2025-06-04T02:19:10
   LastBusyTime=2025-06-04T11:42:38 ResumeAfterTime=None
   CfgTRES=cpu=128,mem=496G,billing=128
   AllocTRES=
   CurrentWatts=96 AveWatts=93