The wiki is deprecated and due to be decommissioned by the end of September 2022.
The content is being migrated to other supports, new updates will be ignored and lost.
If needed you can get in touch with EGI SDIS team using operations @ egi.eu.
Difference between revisions of "GPGPU-WG KnowledgeBase"
Jump to navigation
Jump to search
| Line 16: | Line 16: | ||
* Glue 1.3 Implementation | * Glue 1.3 Implementation | ||
** Based on ideas for supporting "Short Dealine Jobs" and MPI-WG | ** Based on ideas for supporting "Short Dealine Jobs" and MPI-WG | ||
** Proposes establishing specific batch queue following a naming convention, and uses SoftwareEnvironment tags to publish basic software support or hardware details. | ** [[GPGPU-WG:GPGPU_Working_Group_KnowledgeBase:Batch_Schedulers:SchedulerScenarios | Proposes establishing specific batch queue following a naming convention]], and uses SoftwareEnvironment tags to publish basic software support or hardware details. | ||
** Usage of resources not handled well, unless the GPGPUs are exposed through a distinct CE or SubCluster | ** Usage of resources not handled well, unless the GPGPUs are exposed through a distinct CE or SubCluster | ||
** WMS Ranking Support(?) | ** WMS Ranking Support(?) | ||
Revision as of 22:52, 28 February 2014
This page indexes the current knowledge-base on GPGPU integration into the EGI Grid Infrastructure.
- Glue 2.0
- Glue 1.3 Implementation
- Based on ideas for supporting "Short Dealine Jobs" and MPI-WG
- Proposes establishing specific batch queue following a naming convention, and uses SoftwareEnvironment tags to publish basic software support or hardware details.
- Usage of resources not handled well, unless the GPGPUs are exposed through a distinct CE or SubCluster
- WMS Ranking Support(?)