HPC scheduled outages for disk optimisation
Sunday 22 Nov, 6:00pm - Compute service outage commencesMonday 23 Nov, 8:00am - Storage service outage commences Monday 23 Nov, 12:00pm - Outage completes
Sunday 22 Nov, 6:00pm - Compute service outage commences
Monday 23 Nov, 8:00am - Storage service outage commences
Monday 23 Nov, 12:00pm - Outage completes
The wivenhoe cluster uses torque and the maui scheduler. Some options that you would be used to from PBSPro do not work, but have equivalents.
A useful summary of the options for selecting resources for your job can be found at
http://www.clusterresources.com/torquedocs21/2.1jobsubmission.shtml#resources
For example, to use 8 cpus(cores) on 6 nodes you would request:
-l nodes=6:ppn=8
To request 4GB of RAM for you job you would say
-l mem=4gb
To request a wall time of 48 hours, 32 minutes and 10 seconds, you would say
-l walltime=48:32:10