Note - SLURM uses the word partition, but to avoid confusion for people used to using other queueing systems, we use the word queue.
Before creating and submitting jobs, it’s useful to understand the concept of queues, what queues are available for use, and the differences between them.
Queues form part of the cluster functionality provided by the scheduler (SLURM), the product used to balance utilisation across the ADA cluster.
A queue is a pre-defined collection of nodes of a specific type, generally tailored for running a specific type of job. So your first task before running a job is to decide which queue it should be run on.
You need to specify the time for your job - this is different from hpc.uea.ac.uk.
If you don't you will be allocated the default job length of 24 hours.
The maximum job length is 7 days (168 hours). Jobs exceedig this will be killed automatically.
The queues on ADA are as follows:
|Queue Name||Slots Available||Default Time||Maximum Time||RAM Per Core||Priority||Description|
|compute||1200||24 hours||7 days||4GB||30||standard compute node|
(2 per node)
|24 hours||7 days||192GB||30||GPU|
|hmem||96||24 hours||7 days||32GB||20||High memory|
|ib||672||24 hours||7 days||4GB||30||parallel ib mellanox|
|sky||384||24 hours||7 days||4GB||30||skylake compute node|
|broad||192||24 hours||7 days||4GB||30||broadwell compute node|