Skip to content

Free Access

Hive provides free/sponsored access to a limited amount of resources and well as access to otherwise unused compute and GPU resources. Farm only allows access to otherwise unused compute and GPU resources.

See here for instructions to request access to the free resources on Farm and Hive.

Hive

Partition high

In the high partition, every member of publicgrp has combined access to 128 CPUs, 2000 GB of RAM, and 4 GPUs (NVIDIA A6000s).

  • Each job is limited to a maximum of:

    • 8 CPUs
    • 128 GB of RAM
    • 1 GPU

  • You can access these resources by adding --account=publicgrp --partition=high to your srun or sbatch command.

Farm and Hive

GPUs

All GPUs in both clusters are available for use when not otherwise used by the PIs who paid for them. The partition low provides access to currently idle GPUs.

  • Each job is limited to a maximum of 7 days (7-00) of runtime.

  • You can access these resources by adding --partition=low --gpus=1 to your srun or sbatch command.

  • If you require a specific type of GPU, you can request it with --gpus=TYPE:1. You can run this command to see the types and counts of GPUs on each cluster: sinfo -p low -o "%G|%D" | column -s'|' -t

Low partition

The low partition allows access to all the unused resources in the cluster. The downside is that when a job is submitted to high that needs these resources, your job will be killed and requeued. Unless your job can do automatic check-pointing and restarting, all progress is lost when the job is killed.

  • Each job is limited to a maximum of 7 days (7-00) of runtime.

  • You can access these resources by adding --account=publicgrp --partition=low to your srun or sbatch command.

Every user is automatically a member of publicgrp and has access to the low partition.