High Performance Computing Systems
Onyx
Node Configuration
Onyx is a Cray XC40/50 system. It has 4,806 standard compute nodes, 8 large-memory compute nodes, 32 GPU compute nodes, 32 Knights Landing (Phi) compute nodes, and 64 Machine Learning Accelerator (MLA) multi-GPGPU nodes (a total of 4,942 compute nodes or 217,128 compute cores). It is rated at 6.06 peak PFLOPS.
Login | Standard | Large-Memory | KNL | GPU | 2-MLA | 10-MLA | |
---|---|---|---|---|---|---|---|
Total Nodes | 12 | 4,806 | 8 | 32 | 32 | 60 | 4 |
Processor | Intel E5-2699v4 Broadwell | Intel E5-2699v4 Broadwell | Intel E5-2699v4 Broadwell | Intel 7230 Knights Landing | Intel E5-2699v4 Broadwell | Intel 6148 Skylake | Intel 6148 Skylake |
Processor Speed | 2.8 GHz | 2.8 GHz | 2.8 GHz | 1.3 GHz | 2.8 GHz | 2.4 GHz | 2.4 GHz |
Sockets / Node | 2 | 2 | 2 | 1 | 1 | 2 | 2 |
Cores / Node | 44 | 44 | 44 | 64 | 22 | 40 | 40 |
Total CPU Cores | 528 | 211,464 | 352 | 2,048 | 704 | 2,400 | 160 |
Usable Memory / Node | 247 GB | 121 GB | 1 TB | 90 GB | 247 GB | 172 GB | 735 GB |
Accelerators / Node | None | None | None | None | 1 | 2 | 10 |
Accelerator | n/a | n/a | n/a | n/a | NVIDIA P100 PCIe 3 | NVIDIA V100 SXM2 | NVIDIA V100 PCIe 3 |
Memory / Accelerator | n/a | n/a | n/a | n/a | 16 GB | 16 GB | 32 GB |
Storage on Node | None | None | None | None | None | None | None |
Interconnect | Ethernet | Cray Aries | Cray Aries | Cray Aries | Cray Aries | InfiniBand | InfiniBand |
Operating System | SLES | CLE | CLE | CLE | CLE | CentOS | CentOS |
Path | Formatted Capacity | File System Type | Storage Type | User Quota | Minimum File Retention |
---|---|---|---|---|---|
/p/home ($HOME) | 897 TB | Lustre | HDD | 100 GB | None |
/p/work ($WORKDIR) | 13 PB | Lustre | HDD | None | 30 Days |
/p/cwfs ($CENTER) | 3.3 PB | GPFS | HDD | 100 TB | 120 Days |
/p/app ($PROJECTS_HOME) | 897 TB | Lustre | HDD | None | None |