Rivanna Maintenance: May 30, 2023

You may continue to submit jobs until the maintenance period begins, but if the system determines your job will not have time to finish, it will not start until Rivanna is returned to service.

All systems are expected to return to service by 6 a.m. on Wednesday, May 31.


Five RTX3090 nodes (4 GPU devices each) have been added to the gpu partition - use --gres=gpu:rtx3090 in Slurm script.


  1. The toolchains gompic gcccuda goolfc will be removed from Rivanna during the maintenance period, since we now have CUDA-aware toolchains based on gcc/11.2.0. If you need assistance with rebuilding your own code, please contact us here. The following affected modules have been migrated:
Module Previous toolchain Replacement
gpunufft/2.1.0 gcccuda/9.2.0_11.0.228 same version under gcc/11.2.0
gromacs/2021.2 goolfc/9.2.0_3.1.6_11.0.228 2022.4 under goolf/11.2.0_4.1.4
mumax3/3.10 gcccuda/9.2.0_11.0.228 same version under gcc/11.2.0

The following modules will be removed from Rivanna during the maintenance period.

Module Removed version Replacement
alphafold* 2.2.0 2.1.2, 2.2.2, 2.3.0
awscli 2.4.12 2.9.17
cellprofiler* 3.1.8 4.2.5
gurobi 9.5.0 10.0.1
imagemagick 7.0.7-0 7.1.0-57
pytorch* 1.10.0 1.8.0, 1.12.0
rapidsai* 21.10 23.02
tensorflow* 2.4.1, 2.8.0 2.7.0, 2.10.0

*Archived containers can be found in /share/resources/containers/singularity/archive.

  1. Upgraded modules:

    • clang/15.0.7

    Default version changes:

    • nextflow/
  2. New modules:

    • bazel/6.1.1
    • clara-parabricks/4.0.3
    • gpumd/3.7
    • kubectl/1.26.2
    • optix/6.5.0, 7.3.0