Skip to content

VSC-4 partitions and QoS

On VSC-4, nodes of the same type of hardware are grouped in to partitions. The quality of service (QoS), formerly called Queue defines the maximum run time of a job and the number and type of allocate-able nodes.

Partitions

Nodes of the same type of hardware are grouped to partitions. There are three basic types of compute nodes, all with the same CPU, but with different amount of memory: 96 GB, 384 GB and 768 GB.

These are the partitions on VSC-4:

Partition Nodes Architecture CPU Cores per CPU (physical/with HT) GPU RAM Use
skylake_0096 702 Intel 2x Xeon Platinum 8174 24/48 No 96 GB The default partition
skylake_0384 78 Intel 2x Xeon Platinum 8174 24/48 No 384 GB High Memory partition
skylake_0768 12 Intel 2x Xeon Platinum 8174 24/48 No 768 GB Higher Memory partition

Type sinfo -o %P on any node to see all the available partitions.

For the sake of completeness there are internally used special partitions, that can not be selected manually:

Partition Description
login4 login nodes, not an actual slurm partition
rackws4 GUI login nodes, not an actual slurm partition
_jupyter reserved for the jupyterhub

Quality of service (QoS)

The QoS defines the maximum run time of a job and the number and type of allocate-able nodes.

The following QoS are available for all normal (=non private) projects:

QoS name Gives access to Partition Hard run time limits Description
skylake_0096 skylake_0096 72h (3 days) Default
skylake_0384 skylake_0384 72h (3 days) High Memory
skylake_0768 skylake_0768 72h (3 days) Higher Memory

Tip

The QoSs that are assigned to a specific user can be viewed with:

sacctmgr show user `id -u` withassoc format=user,defaultaccount,account,qos%40s,defaultqos%20s

Idle QoS

If a project runs out of compute time, jobs of this project are now running with low job priority and reduced maximum run time limit in the idle QoS.

QoS name Gives access to Partition Hard run time limits Description
idle_0096 skylake_0096 24h (1 day) Projects out of compute time
idle_0384 skylake_0384 24h (1 day) Projects out of compute time
idle_0768 skylake_0768 24h (1 day) Projects out of compute time

Devel QoS

The devel QoS gives fast feedback to the user when their job is running. We recommend this before sending the job to one of the compute queues.

QoS name Gives access to Partition Hard run time limits
skylake_0096_devel 5 nodes on skylake_0096 10min

Private Projects

Private projects come with different QoS

QoS name Gives access to Partition Hard run time limits
p7XXXX_YYYY various - check with your project manager up to 240h (10 days)

where 7XXXX is the project number, and YYYY the RAM size.