Translations:Running jobs/106/en: Difference between revisions
Jump to navigation
Jump to search
(Importing a new version from external source) |
(Importing a new version from external source) Tags: Mobile edit Mobile web edit |
||
| Line 1: | Line 1: | ||
Memory may be requested with <code>--mem-per-cpu</code> (memory per core) or <code>--mem</code> (memory per node). We recommend that you specify memory in megabytes (e.g. 8000M) rather than gigabytes (e.g. 8G). In many circumstances specifying memory requests in thousands of megabytes will result in shorter queue wait times than specifying gigabytes. For example, requesting <code>--mem=128G</code> (equivalent to 131072M) is more memory than is available for jobs on nodes with a nominal 128G . A job requesting <code>--mem=128G</code> qualifies for fewer nodes than one requesting <code>--mem=128000M</code>, and is therefore likely to wait longer. Similar reasoning also applies to <code>--mem-per-cpu</code> requests. | |||
Revision as of 14:18, 21 June 2018
Memory may be requested with --mem-per-cpu (memory per core) or --mem (memory per node). We recommend that you specify memory in megabytes (e.g. 8000M) rather than gigabytes (e.g. 8G). In many circumstances specifying memory requests in thousands of megabytes will result in shorter queue wait times than specifying gigabytes. For example, requesting --mem=128G (equivalent to 131072M) is more memory than is available for jobs on nodes with a nominal 128G . A job requesting --mem=128G qualifies for fewer nodes than one requesting --mem=128000M, and is therefore likely to wait longer. Similar reasoning also applies to --mem-per-cpu requests.