For a comparison of ATS-4 systems, see: Using El Capitan Systems: Hardware Overview
| Zone |
RZ
|
| Vendor |
HPE Cray
|
| User-Available Nodes |
Login Nodes*
2
Batch Nodes
102
Debug Nodes
24
Total Nodes
128
|
| APUs |
APU Architecture
AMD MI300A
|
| CPUs |
CPU Architecture
4th Generation AMD EPYC
Cores/Node
96
Total Cores
12,288
|
| GPUs |
GPU Architecture
CDNA 3 [APU: AMD MI300A]
Total GPUs
512
GPUs per compute node
4
GPU global memory (GiB)
128.00
|
| Memory Total (GiB) |
65,536
|
| CPU Memory/Node (GiB) |
512
|
| Peak Performance |
Peak PFLOPS (CPUs)
0.358
Peak PFLOPs (GPUs)
31.744
Peak PFLOPS (CPUs+GPUs)
32.102
|
| Clock Speed (GHz) |
3.7
|
| OS |
TOSS 4
|
| Interconnect |
HPE Slingshot 11
|
| Scheduling Policy (main batch queue) |
node-scheduled
|
| Scheduler |
Flux
|
| Recommended location for parallel file space |
/p/lustre{...}
|
| Program |
ASC
|
| Class |
ATS-4
|
| Year Commissioned |
2024
|
| Compilers |
See Compilers page |
