LAST UPDATE 4th September 2019
Below is a table describing each of the partitions available on Viking. These may change as we come to understand the workloads running on Viking and update these queues to reflect that.
|Queue name||Max time||Max jobs (running)||Max Cores||Default mem per CPU||Max memory||Default time||Nodes in Partition||Comment|
|Nodes||48 hrs||1000||1000||4.8 GB||4.8 TB||8 hrs||170||Default partition for everything.|
2 (4 jobs can be queued)
|8||1 GB||16 GB||10 mins||170||Partition for testing that jobs will start, need to request using –partition=test in job script.|
|himem||48 hrs||No limit||96||11.875 GB||N/A||8 hrs||3||Partition for running jobs that require memory greater than that available in other partitions. Need to request using –partition=himem in job script.|
Two nodes have 64 cores, one node has 96 cores.
|GPU||48 hrs||No limit||80||9.6 GB||N/A||8 hrs||2|
Partition for running jobs that require GPUs, see documentation for details about how to request GPUs.
|Week||7 days||No limit||400||4 GB||N/A||48 hrs||10||Partition for jobs that need to run for longer than the 48h max time allowed by the nodes partition, but for less than a week. Need to request using –partition=week in job script|
|Month||30 days||No limit||120||4 GB||N/A||7 days||3|
Partition for jobs that need to run for longer than the 7 day max time allowed by the week partition, but for less than 30 days. Need to request using –partition=month in job script