cascades1 going offline; switch to cascades2

[ARC]

Cascades Users,
The cascades1 login node will be going offline for maintenance in a few minutes. Please use the cascades2 login node for submitting jobs to the V100 GPU nodes from now on; for more on the V100 nodes, please see below. Please also note that we are continuing to work through maintenance on the Cascades cluster; maintenance is expected to continue into next week.

We appreciate your patience while this work on Cascades is completed.
Regards,
Advanced Research Computing
http://www.arc.vt.edu



On Thursday, April 12, 2018 at 2:59:51 PM UTC-4, Advanced Research Computing wrote:
ARC Users,

ARC is happy to announce the addition of 39 new GPU nodes to the Cascades cluster, which are now online and accessible via cascades1.arc.vt.edu even as the outage on the other Cascades nodes is ongoing. Each of the new nodes is equipped with two Intel Skylake Xeon Gold 3 GHz CPU (24 cores/node in all), 384 GB memory, and two NVIDIA V100 GPUs. Each Volta GPU is capable of greater than 7.8 TeraFLOPS of double-precision performance, so including CPU and GPU these nodes add over 800 TFLOPS of peak double-precision throughput to ARC\\\'s resources.

Usage notes:

- Please restrict work on these nodes to GPU-enabled applications, or applications that require the Skylake architecture.
- The V100 nodes are named ca197 through ca236.
- The CPU architecture on these nodes (Skylake) is newer than on the (Broadwell) login nodes, so you should compile your code on the compute nodes for maximum performance.
- We have created two queues for these nodes:
- v100_normal_q for production jobs
- v100_dev_q for code development and testing
- You can start an interactive job on the nodes with this command:
interact -q v100_dev_q -lnodes=1:ppn=24:gpus=2

Please let us know if you have questions or run into problems getting started.

Advanced Research Computing
http://www.arc.vt.edu