Dear DICC Users,
We are pleased to announce that 4 more compute nodes have been brought online in the HPC cluster:
- 3 Opteron CPU nodes that get added back to the cpu-opteron partition.
- 1 new GPU server, which comes with 2 x Tesla v100s GPU cards, 64 CPUs and 192 GB Memory. This node will be associated with the gpu-v100s partition, which has similar applications installed as other GPU servers. Our quick benchmark results indicate that v100s is at least 10 times faster than Titan Xp.
Users may start to utilise the new added compute machine to run your jobs. If you encounter any problem trying to run a job on those machines, please let us know through the service desk.
Thank you.