Body
On the Mill there are several partitions that you can submit jobs to. The optimal partition will depend on your resource needs. A summary of the various partitions is below, with more details on the individual partitions following.
| |
Maximum Time Limit |
Description |
| priority partitions |
28 days |
For investors. |
| requeue |
2 days |
For non-investor jobs that have been requeued due to their landing on an investor-owned node. |
| general |
2 days |
For non-investors to run multi-node, multi-day jobs. |
| gpu |
2 days |
Acceptable use includes jobs that utilize a GPU for the majority of the run. Is composed of Nvidia A100 cards, 4 per node. |
| interactive |
4 hours |
For short interactive testing, interactive debugging, and general interactive jobs. Use this for light testing as opposed to the login node. |
| class |
4 hours |
This partition is for students working on coursework. |
Lab Partitions
Up to 50% of the total compute capacity in the Mill can be leased by individual research groups. If your lab has purchased or leased computing resources in the Mill, then you will have access to a lab partition (commonly named as your PIs's last name). Submitting your jobs to this partition will generally offer the best performance.
Requeue
The requeue partition is the default partition. This is where jobs will be queued if no partition is set in the SLURM submission script. When leased/purchased nodes are idle they will accept jobs from the requeue partition. These jobs will be interrupted and sent back to the queue should the owner of the node submit a job, and the owner's job will start immediately. This makes requeue an ideal partition for short jobs that are unlikely to be interrupted before they finish and jobs with robust checkpointing that can make non-continuous progress in the case of an interruption.
General
The general partition will have access to similar hardware as the requeue partition, but in exchange for potentially longer wait times, jobs will not be interrupted once started.
GPU
The gpu partition should be used for jobs that require GPU resources. When submitting jobs to the gpu partition, please keep in mind that GPU resources are limited and GPU acceleration is very popular. All users should take care to reasonable optimize their job's performance and request only the resources that their job can use. Failure to do so degrades the quality of service for all users.
Interactive
The interactive partition is intended for interactive sessions rather than asynchronous jobs. Good use cases include things like graphical output for visualizations, rapid prototyping, and debugging.
Class
The class partition is used for student coursework. This is the only partition that students will have access to.