Dear colleagues
I am not a bioinformatician. I have access to the bigmem partition.
Here in the beginning of .sh
#!/bin/bash
#SBATCH --job-name=paresnip2
#SBATCH --ntasks=1
#SBATCH --mem-per-cpu=30000
#SBATCH --cpus-per-task=12
#SBATCH --partition=long
#SBATCH -o paresnip2_testdata.out
The error is
sbatch: error: Batch job submission failed: Requested node configuration is not available
Please someone can help me?
Kind regards
Julie
HI Julie
You can refer to our documentation for help: Slurm user guide - IFB Core Cluster Documentation
Currently you are submitting to the long partition, and not to bigmem
#SBATCH --partition=long
Thank you
Nicole
Hello @nc-support ,
I followed the slurm user guide to submit the job, but it seems error with CUDA configuration; I checked the available modules, but CUDA is not available. Is there any way to solve this issue?
The error I am getting is as follows:
CUDA initialization error on cpu-node-21.ifb.local: CUDA driver version is insufficient for CUDA runtime version
FATAL ERROR: CUDA error cudaGetDeviceCount(&deviceCount) in file src/DeviceCUDA.C, function initialize, line 142
on Pe 0 (cpu-node-21.ifb.local): CUDA driver version is insufficient for CUDA runtime version
Thanks in advance!
Hello Hrishikesh,
To use GPU acceleration with the CUDA driver, you must use a node with GPU cards.
To use it, once you have access (it seems ok: Acces à la demande à la partition gpu - #3 par nc-support), you must specify the GPU partition: -p gpu
Please, to facilitate the support, open a new subject for each new request.
Best regards,