Page History
Gadi Cascade Lake node = 48 CPUs, 192 GB memory
1 cpu-hour = 2 service units (SU)Job cost = ncpus_request x max[ 1, (ncpus_per_node/mem_per_node)x(mem_request/ncpus_request)] x walltime_usage x queue_charge_rate
Queue | CPUs request | GPUs request | Memory Request (GB) | Walltime Usage |
|---|
Cost | Comments |
|---|
| normal | 4 | n.a. | 16 GB | 5 hours | 4 x 5 x 2 = 40 SU |
| normal | 8 | n.a. |
| 16 GB | 5 hours | 8 x 5 x 2 = 80 SU | CPU request dominates. |
| normal | 8 | n.a. | 128 GB | 5 hours |
| 48 x (128/192) x 5 x 2 = 320 SU | Memory request dominates. |
32
| 32=48*(128/192) cpus is proportion of node resources. |
| normal | 8 | n.a. | 192 GB | 5 hours | 48 x 5 x 2 = 480 SU | Memory request dominates. |
192GB = 100% of node memory.
| express | 8 | n.a. |
| 16 GB | 5 hours | 8 x 5 x |
| 6 = 240 SU | Express charge rate is 6.0 | ||||
| gpuvolta | 12 | 1* | 90GB | 5 hours | 12 x 5 x 3 = |
| 180 SU | CPU request dominates |
Express multiplier is x3.
| . *Because of the `ncpus = 12 x ngpus` constraint in GPU job request, ngpus determines the cost through ncpus in the equation. | ||||||
| gpuvolta | 12 | 1* | 380GB | 5 hours | 48 x (380/382) x 5 x 3 = 716.23 SU | Memory request dominates. |
Overview
Content Tools