Revision as of 12:51, 29 April 2016

Default Execution Limits

Execution limits at the Centre for Advanced Computing are in place to provide greater flexibility in scheduling user applications. The following table shows the default limits that are currently in place. Note that our clusters are undergoing extensions and re-structuring, and that these limits are likely to change in the near future. Please check sporadically to keep up-to-date with them.

• Global Job Limit: Users are limited to 48 executing jobs at any one time. This means that across all clusters each user can run up to 48 (total) production jobs simultaneously, serial or parallel. Submitted jobs above that limit will remain in the queue until a job slot comes free.

• Process or Thread Limit: A maximum number of processes or threads that can be run at a given time is in place to cap standard jobs at the default allocation for a given cluster. The current value for the SW (Linux) Cluster (sw0011-56) is 48.

Important Note: Research groups will in many cases require more resources, especially in terms of thread/process numbers, and may fall outside of these limits. We will work with these users to ensure appropriate access. Any requests for enhanced access should include specifications for number of processors required, total amount of memory, and the expected maximum runtime(s). Contact the User Support group to ensure that this is arranged using the most appropriate resource.

Long-term extended usage requires a formal application to Compute Canada. Calls for such applications are currently issued once a year in the fall. We are announcing these calls on our web page. The allocation is done by a Research Allocation Committee (RAC).

Licensed software packages such as Fluent and Abaqus may be subject to additional limits. Such limits are presently suspended but may be re-instituted if the limit of available licenses is approached.