Image
| Zone |
CZ
|
| Vendor |
IBM
|
| User-Available Nodes |
Login Nodes*
2
Batch Nodes
54
Total Nodes
62
|
| CPUs |
CPU Architecture
IBM Power8
Cores/Node
20
Total Cores
1,240
|
| GPUs |
GPU Architecture
NVIDIA Tesla P100
Total GPUs
216
GPUs per compute node
4
GPU peak performance (TFLOP/s double precision)
1,015.00
GPU global memory (GB)
3456.00
|
| Memory Total (GB) |
17,280
|
| CPU Memory/Node (GB) |
256
|
| Peak Performance |
Peak TFLOPS (CPUs)
25.2
Peak TFLOPs (GPUs)
1,015.2
Peak TFLOPS (CPUs+GPUs)
1,040.40
|
| Clock Speed (GHz) |
4.0
|
| Peak single CPU memory bandwidth (GB/s) |
115
|
| OS |
RHEL
|
| Interconnect |
IB EDR
|
| Parallel job type |
multiple nodes per job
|
| Recommended location for parallel file space |
/p/gscratchr
|
| Program |
ASC, M&IC
|
| Year Commissioned |
2017
|
| Compilers | |
| Documentation |