SLURM at IFB Core
The Partitions and resource limits
To request access to partitions on demand such as the bigmem node or the gpu nodes, please submit your request on the community support platform: community support website specifying your login and project name.
⚠️ The values below can change. To check the current situation:
scontrol show partition
sacctmgr list qos format=Name,Priority,MaxTRESPU%20
| Partitions | Time out | Max resources / user | Purpose |
|---|---|---|---|
fast |
<= 24 hours | cpu=300, mem=1500GB | Default - Regular jobs |
long |
<= 30 days | cpu=300, mem=1500GB | Long jobs |
bigmem |
<= 60 days | mem=4000GB | On demand - For jobs requiring a lot of RAM |
gpu |
<= 3 days | cpu=300, mem=1500GB | On demand - Access GPU cards |
The default values
| Param | Default value |
|---|---|
--mem |
2GB |
--cpus |
1 |
IFB Core Cluster Computing nodes
⚠️ The values below can change. To check the current situation:
sinfo -Ne --format "%.15N %.4c %.7z %.7m" -S c,m,N | uniq
(Last update: 04/05/2022)
CPU nodes
| Nbr | CPU (Hyper-Thread) | RAM (GB ) |
|---|---|---|
| 10 | 30 | 128 |
| 1 | 38 | 256 |
| 66 | 54 | 256 |
| 1 | 124 | 3096 |
GPU nodes
| Nbr | CPU | RAM (GB) | GPU | Type | Disk /tmp |
|---|---|---|---|---|---|
| 3 | 62 | 515 | 2 | NVIDIA Ampere A100 40GB | 4 TB |
GPU Cards have been partitioned into isolated GPU instances with Multi-Instance GPU (MIG).
GPU Instance Profiles
⚠️ The values below can change. To check the current situation:
sinfo -Ne -p gpu --format "%.15N %.4c %.7m %G"
| Profile Name | GPU Memory | GPU Compute Slice | Number of Instances Available |
|---|---|---|---|
| 1g.5gb | 5GB | 1 | 14 |
| 3g.20gb | 20GB | 3 | 2 |
| 7g.40gb | 40GB | 7 | 3 |