This cluster has been decommissioned in July 2018.
Hosted at UCLouvain, Lemaître2 consisted of 112 compute nodes, each equipped with
two 6-core Intel E5649 @ 2.53 GHz processors and 48 GB RAM (4 GB per core).
The cluster had exclusive access to a high-performance 120 TB Lustre parallel filesystem,
and all nodes (compute, NFS, Lustre, frontend…) were interconnected via QDR Infiniband.
Suitable for:
Massively parallel jobs (MPI, dozens of cores) requiring high communication throughput
and/or heavy parallel disk I/O.
Max wall time: 3 days.
Resources
Home directory (50 GB per user)
Working directory
/scratch($GLOBALSCRATCH)Internet access available on compute nodes
Queues*:
- Default queue — 3 days max, 50 running jobs/user
- PostP queue (with GPUs) — 6 hours
Generic resource*:
gpu
Access / Support
This cluster has been decommissioned and is no longer accessible.
