Deneb is an Intel Xeon based cluster available to the EPFL comunity since November 17th, 2014.
- Peak performance: 293 TFLOPs (211 in CPUs, 92 in GPUs)
- Total RAM: 37TB
- Storage: 350TB
The cluster is composed of
- two front ends,
- two master nodes in charge of administration and backup,
- 376 compute nodes, each with
- 2 Ivy Bridge processors running at 2.6 GHz, with 8 cores each,
- 64 GB of DDR3 RAM,
- 144 compute nodes, each with
- 2 Haswell processors running at 2.5 GHz, with 12 cores each
- 64 GB of DDR4 RAM
- 16 GPU accelerated nodes, each with 4 K40 NVIDIA cards
- 8 large memory nodes, each with 256 GB of RAM
- 2 NUMA nodes, each with 4 processors and 512 GB of RAM
- Infiniband QDR 2:1 connectivity,
- GPFS filesystem.
Requesting access to the GPU nodes
In order to use the GPU nodes we request that you submit a description of the code you wish to use and the performance benefits expected. You will then be invited to meet our application and GPU experts to discuss your proposal.