Juggernaut Cluster
The Juggernaut HPC cluster was a small cluster maintained by DIT to allow the needs of some researchers until the Zaratan cluster came online. The Juggernaut cluster was retired in July 2024 and is no longer available. This section is for historical purposes only.
| Description | Number of nodes | Processor | Cores/node | Mem/node (GiB) | Mem/core | GPUs/node |
|---|---|---|---|---|---|---|
| green partition | 13 | Skylake | 40 | 384 | 9.6 | none |
| blue partition | 4 | Broadwell | 28 | 256 | 9.1 | none |
| green part GPU | 1 | CascadeLake | 80 | 1536 | 19 | V100 |
| blue part GPU | 1 | SkyLake | 32 | 384 | 12 | P100 |
The standard compute nodes in the green partition have EDR infiniband interconnects, and dual Intel Xeon Gold 6148 Skylake processors running at 2.4 GHz with 20 cores per CPU.
The standard compute nodes in the blue partition have dual Intel E5-2680v4 Broadwell EP processors running at 2.4 GHz with 14 cores per CPU.
The green partition GPU node has quad Intel Xeon Gold 6248 CascadeLake processors running at 2.5 GHz with 20 cores per CPU. It also hosts a single NVIDIA Tesla Volta V100 GPU.
The blue partition GPU node has quad Intel Xeon Gold 6142 SkyLake processors running at 2.6 GHz with 16 cores per CPU. It also hosts a single NVIDIA Tesla Pascal P100 GPU.
Note
The Juggernaut cluster was retired in 2024. It is no longer available for computation. The information in this page is for historical records only.