This cluster has been decommissioned in August 2019.
Hosted at UNamur, the Hercules cluster consisted of 512 cores across 32 Intel Sandy Bridge nodes, each equipped with two 8-core E5-2660 @ 2.2 GHz CPUs and 64 GB RAM (or 128 GB RAM on 8 nodes).
All nodes were connected via Gigabit Ethernet and had access to three NFS filesystems providing a combined 100 TB of storage.
Suitable for:
Long shared-memory parallel jobs (OpenMP, Pthreads) or resource-intensive sequential workloads.
Max wall time: 63 days.
Resources
- Home directory (200 GB per user)
- Working directory
/workdir($WORKDIR, 400 GB per user) - Local node storage
/scratch($TMPDIR, dynamically assigned in jobs) - No internet access from compute nodes
- CPU queue* (Max 63 days, 48 CPUs/user)
Access / Support
This cluster has been decommissioned and is no longer accessible.
