Enabling UB's Research Computing Since 1998
Industry Compute Partition
New Compute Nodes Available December 2024
The compute hardware available to industry partners includes two sets of purchases from 2021 and 2024. These equipment purchases were funded by the Buffalo Institute for Genomics and a New York State Empire State Development grant to provide the WNY and NY State industrial community with access to state-of-the-art high-performance computing resources (hardware, software and consulting services) to help foster economic development.
The 2021 equipment consists of 99 Dell PowerEdge servers with a total of 5544 processor cores and HDR InfiniBand interconnect. 67 of these nodes consist of two Intel Ice Lake Xeon Gold 6330 28-core processors, 512GB of memory and 960GB of local scratch. Another 16 of the nodes consist of two Intel Ice Lake Xeon Gold 6330 28-core processors, 1024GB of memory and 960GB of local scratch. The final 16 nodes consist of two Intel Ice Lake Xeon Gold 6330 28-core processors, dual Nvidia A100 GPU, 512GB of memory and 960GB of local scratch. Currently, there are high-bandwidth memory compute nodes and NVIDIA DGX 8-way H100 nodes available from the 2024 equipment purchase. Additional compute nodes from this purchase and their full details will be available soon.
Server (Node) Types:
Type of Node | # of Nodes | # CPUs | Processor | GPU | RAM | Network | SLURM TAGS | Local /scratch |
| "Ice Lake" Standard Compute Node | 67 | 56 | Intel Xeon Gold 6330 | - | 512GB | HDR Infiniband | CPU-Gold-6330 | 960GB |
"Ice Lake" Large Memory | 16 | 56 | Intel Xeon Gold 6330 | - | 1024GB | HDR Infiniband | CPU-Gold-6330 | 960GB |
"Ice Lake" GPU Node | 16 | 56 | Intel Xeon Gold 6330 | Dual Nvidia A100 | 512GB | HDR Infiniband | CPU-Gold-6330-A100 | 960GB |
| "Sapphire Rapids" Standard Compute Node | 75 | 64 | Intel Gold-6448Y | - | 512GB | Infiniband**** | CPU-Gold-6448Y, SAPPHIRE-RAPIDS-IB | 880GB |
| "Sapphire Rapids" Large Memory Node | 12 | 64 | Intel Gold-6448Y | - | 2TB | Infiniband | CPU-Gold-6448Y, SAPPHIRE-RAPIDS-IB | 7TB |
| "Sapphire Rapids" GPU Node | 4 | 64 | Intel Gold-6448Y | H100 | 512GB | Infiniband | CPU-Gold-6448Y, SAPPHIRE-RAPIDS-IB, H100 | 7TB |
| High Bandwidth Memory Node | 8 | 64 | Intel Max 9462 | - | 125GB | HDR Infiniband | CPU-Max-9462, HBM | 833GB |
| NVIDIA DGX GPU Node | 2 | 96 | Intel Platinum 8562Y | 8x H100 | 2TB | HDR Infiniband | CPU-Platinum-8562Y, H100 | 14TB |
Requesting Specific Node Resources:
Partitions Available:
The industry nodes are in three partitions within the UB-HPC cluster: industry, industry-dgx, and industry-hbm. These partitions are only available to industry partners. UB faculty and students have the option of running their jobs in the "scavenger" partition. This allows jobs to run when there are no other pending jobs in the industry partitions. Once an industry user submits a job requesting resources, jobs in the scavenger partition are stopped and requeued.
Note to use the scavenger partition, your jobs MUST be able to checkpoint and restart.
| Partition Name | Time Limit | Max jobs per user | Notes |
| industry | 72 hours | 1000 | contains a mix of standard compute nodes, large memory and GPU nodes |
| industry-dgx | 72 hours | 10 | contains NVIDIA DGX GPU nodes |
| industry-hbm | 72 hours | 10 | contains high-bandwidth memory nodes |
| scavenger | 72 hour | 1 | --requeue flag required for jobs to be restarted |