CLUSTER
The platform ABiMS offers resources around a dedicated data center (Marine Genomic Data Center).
This data center is destined to data storage and security, as well as their analyses.
It represents :
-
2600 Core CPU handled by the scheduler SLURM,
-
a secure data storage (RAID) of 2 PB
This infrastructure has been financed by multiple sources :
-
Région Bretagne - BioGenOuest
-
Conseil Général - CG29
-
ANR projects
-
Station Biologique de Roscoff: FR2424, UMR7139, UMR7144, UMR7150
-
Investissements d'Avenir projects: EMBRC-France, IDEALG, Océanomics, IFB
-
GRSIBI projects.
In 2013, 90% of the CPU resources involved collaborative projects with at less one partner from the "Station Biologique de Roscoff"; 10 % were used for fully external projects.
Training Materials
Please have a look at this Documentation provided by the IFB NNCR Cluster TaskForce before using the Cluster.
With the IFB Core Cluster, we share all our pratices.
Resources
DIFFERENTS TYPES OF NODE
sinfo -Nel --format "%.8N %.9R %.4c %.7z %.7m"
| NODELIST | PARTITION | CPUS | Socket:Core:Thread | MEMORY IN MB |
|---|---|---|---|---|
| n75 | fast | 254 | 2:127:1 | 257445 |
| n82 | fast | 30 | 2:15:1 | 128882 |
| n84 | fast | 30 | 2:15:1 | 128882 |
| n84 | long | 30 | 2:15:1 | 128882 |
| n86 | fast | 30 | 2:15:1 | 128882 |
| n86 | long | 30 | 2:15:1 | 128882 |
| n87 | fast | 30 | 2:15:1 | 112754 |
| n87 | long | 30 | 2:15:1 | 112754 |
| n88 | fast | 30 | 2:15:1 | 128882 |
| n88 | long | 30 | 2:15:1 | 128882 |
| n89 | fast | 30 | 2:15:1 | 128882 |
| n89 | long | 30 | 2:15:1 | 128882 |
| n90 | fast | 30 | 2:15:1 | 128882 |
| n90 | long | 30 | 2:15:1 | 128882 |
| n91 | fast | 30 | 2:15:1 | 128882 |
| n91 | long | 30 | 2:15:1 | 128882 |
| n92 | fast | 30 | 2:15:1 | 128882 |
| n92 | long | 30 | 2:15:1 | 128882 |
| n93 | fast | 30 | 2:15:1 | 128882 |
| n93 | long | 30 | 2:15:1 | 128882 |
| n94 | fast | 30 | 2:15:1 | 128882 |
| n94 | long | 30 | 2:15:1 | 128882 |
| n95 | fast | 30 | 2:15:1 | 128882 |
| n95 | long | 30 | 2:15:1 | 128882 |
| n96 | fast | 254 | 2:127:1 | 257445 |
| n96 | long | 254 | 2:127:1 | 257445 |
| n97 | fast | 254 | 2:127:1 | 257445 |
| n97 | long | 254 | 2:127:1 | 257445 |
| n98 | fast | 126 | 2:63:1 | 257468 |
| n98 | long | 126 | 2:63:1 | 257468 |
| n99 | bigmem | 80 | 4:10:2 | 1032076 |
| n100 | bigmem | 128 | 4:16:2 | 2064169 |
| n115 | fast | 46 | 2:23:1 | 257859 |
| n115 | long | 46 | 2:23:1 | 257859 |
| n117 | fast | 46 | 2:23:1 | 257859 |
| n117 | long | 46 | 2:23:1 | 257859 |
| n118 | fast | 46 | 2:23:1 | 257859 |
| n118 | long | 46 | 2:23:1 | 257859 |
The available partitions are:
scontrol show partition sacctmgr list qos format=Name,Priority,MaxTRESPU%20
| Time out | Default resources | Max resources / user | Purpose | |
|---|---|---|---|---|
| fast | 24 hours | --cpu-per-task 1 --mem 2GB | cpu=200, mem=750GB | Regular jobs |
| long | 30 days | --cpu-per-task 1 --mem 2GB | cpu=200, mem=750GB | Long jobs |
| bigmem | 60 days | --cpu-per-task 1 --mem 2GB | mem=1.9TB | On demande. For treatements requiring a lot of RAM |