Queues (Partitions)
Vista includes CPU and GPU resources to meet a broad range of computational demands. Accordingly, users will submit jobs that differ significantly in resource type, size, duration, and priority. Vista's batch queues—which Slurm refers to as "partitions"—have been configured so that users gain access to their preferred resources fairly and efficiently.
Here is a summary of the characteristics of the different queues on Vista at the time of this writing (October, 2026). A Grace Blackwell queue was recently added, as a kind of preview of Horizon:
| Queue Name | Max Nodes/Job (Cores/Job) |
Max Job Time |
Max Jobs per User |
Max Jobs Submitted |
Max Total Nodes per User |
Charge per Node-Hour |
|---|---|---|---|---|---|---|
| gb | 18 nodes (2,592 cores) |
12 hrs | 2 jobs | 3 jobs | 18 nodes | 1.0 SU |
| gg | 32 nodes (4,608 cores) |
48 hrs | 20 jobs | 40 jobs | 96 nodes | 0.33 SU |
| gh | 64 nodes (4,608 cores/64 GPUs) |
48 hrs | 20 jobs | 40 jobs | 96 nodes | 1.0 SU |
| gh-dev | 8 nodes (576 cores) |
2 hrs | 1 job | 3 jobs | 8 nodes | 1.0 SU |
Jobs may be submitted to any of the above queues. Two of the queues feature GPUs. The nodes in the gh and gh-dev queues feature one Hopper H100 GPU per node. Detailed specifications for all the types of nodes on Vista can be found in the Slurm Partitions section of the Vista User Guide.
Be aware that the queue configuration in the table above is subject to change at any time.
To get up-to-date queue configurations, run this on a login node:
$ qlimits -vCVW material development is supported by NSF OAC awards 1854828, 2321040, 2323116 (UT Austin) and 2005506 (Indiana University)