Skip to content

Vega

EESSI, the base of the EFP Federated Software Catalog, is available system-wide on Vega via a native CernVM-FS deployment since early 2025.

Synchronisation

Vega is synchronised directly via the public EESSI infrastructure. This means that once new software has been integrated into EESSI, the maximum wait time for it to appear on the system is under an hour.

Access

Standard access instructions apply.

Usage

EESSI is available by default on both login and compute nodes but requires initialisation. See the initialisation instructions for the Federated Software Catalog and then continue with loading your required software.

Known problems and limitations

  • None

System-specific best practices

Multi-node jobs

We provide an example of running a latency test between nodes from the OSU Micro-Benchmarks suite. This is done on 2 nodes, each with GPUs, and checking the device-to-device (node-to-node) latency: 

# Allocate interactive resources for the test
# (2 nodes, 1 task per node, 1 GPU per task)
srun --time=00:30:00 --account="<SLURM_ACCOUNT>" --partition=gpu --nodes 2 --ntasks-per-node 1 --cpus-per-task 1 --gres=gpu:1 --pty bash -i
# Configure the EESSI environment and load the modules we need for our test
source /cvmfs/software.eessi.io/versions/2025.06/init/lmod/bash
module load OSU-Micro-Benchmarks/7.5-gompi-2025a-CUDA-12.8.0
# Check we have two different nodes
# (we are already in an srun session so we need to overlap, and disable cpu binding)
srun --mpi=pmix -n 2 --overlap --cpu-bind=none hostname
# Launch our latency test
# (we are already in an srun session so we need to overlap, and disable cpu binding)
srun --mpi=pmix -n 2 --overlap --cpu-bind=none osu_latency --accelerator cuda

The --mpi=pmix is not strictly necessary, we include it above for completeness.

Getting help with using FSC on Vega

If you need help with using EESSI on Vega, please contact EFP Helpdesk.

References