Differences

This shows you the differences between two versions of the page.

--- wiki:uso_sist_en [2022/10/06 20:08]
cnr-guest [Tips for using Matlab for the execution of parallel jobs on the IBiSCo HPC cluster]
+++ wiki:uso_sist_en [2022/10/07 19:14] (current)
cnr-guest [Available file systems]
@@ Line 73: / Line 73: @@
 In-depth documentation on Lustre is available online, at the link: '' https://www.lustre.org/ ''
+''/ibiscostorage''
+new scratch area shared among UI and computation nodes (available from 07/10/2022), **not** LUSTRE based
 ==== Job preparation ans submission ====
+=== Premise: new job management rules active from 9/10/2022 ===
+To improve the use of resources, the job management rules have been changed.
+ * New usage policies based on // fairshare // mechanisms have been implemented \\
+ * New queues for job submissions have been defined
+    - ** sequential ** queue:
+      * accepts only sequential jobs with a number of tasks not exceeding 1,
+      * who do not use GP-GPUs,
+      * for a total number of jobs running on it not exceeding 128
+      * and maximum execution time limit of 1 week
+    - ** parallel ** queue:
+      * accepts only parallel jobs with task number greater than 1 and less than 1580,
+      * that use no more 64 GP-GPUs
+      * and maximum execution time limit of 1 week
+     - ** gpus ** queue:
+      * only accepts jobs that use no more than 64 GP-GPUs,
+      * with task number less than 1580
+      * and maximum execution time limit of 1 week
+     - ** hparallel ** queue:
+      * accepts only parallel jobs with task number greater than 1580 and less than 3160,
+      * that make use of at least 64 GP-GPUs
+      * and maximum execution time limit of 1 day
+From 9 October the current queue will be disabled and only those defined here will be active, to be explicitly selected. For example, to subdue a job in the ** parallel ** queue, execute \\
+  $ srun -p parallel <MORE OPTIONS> <COMMAND NAME>
+If the job does not comply with the rules of the queue used, it will be terminated.
+=== Use of resources ===
 In the system is installed the resource manager SLURM to manage the cluster resources.

ibisco HPC Wiki

User Tools

Site Tools

Differences

Page Tools