The RES periods are divided into three, each with a duration of 4 months:
The different resource limits that RES users will have are associated to different QOS (Quality Of Services). The QOS associated to the SCAYLE nodes are:
Name | Priority | MaxWall* |
---|---|---|
class_a | 100 | 3-00:00:00 |
class_b | 10 | 2-00:00:00 |
class_c | 1 | 1-00:00:00 |
*MaxWall: This is the maximum time that can be requested when submitting a job (days-hours:minutes:seconds).
At the beginning of each RES period, a QOS will be assigned to each user so that they can perform their calculations. If, during the period, a user reaches the limit of hours assigned for the current period, from that moment on he/she will have to change the sending QOS. This will be notified by our Technical Dept. by sending an email.
Here we can see a simple example of a script with the parameters we need to launch it.
#!/bin/bash
#SBATCH --job-name=<JOBNAME>
#SBATCH --output <my_job.out>
#SBATCH --error <my_job.err>
#SBATCH --partition=res
#SBATCH --qos=class_a
#SBATCH --ntasks-per-node=16
#SBATCH --time=03:00:00
#SBATCH --mail-user=<EMAIL>
#SBATCH –-mail-type=ALL
# Your job below
srun --label hostname
For more information about the script go to the Job Manager section of the manual.
You can submit the created script using the command:
sbatch job.sh
Last update: 06/02/2025 13:30