In coordination with the ACCESS team, XSEDE has paused the processing of Startup, Education, and other allocation requests from August 17-31. This pause will ensure that no requests are lost while we are making infrastructure updates behind the scenes and handing things over to ACCESS. If you have questions, contact help@xsede.org. The ACCESS team will resume taking allocation requests on September 1 via https://access-ci.org/.

The discussion forums in the XSEDE User Portal are for users to share experiences, questions, and comments with other users and XSEDE staff. Visitors are welcome to browse and search, but you must login to contribute to the forums. While XSEDE staff monitor the lists, XSEDE does not guarantee that questions will be answered. Please note that the forums are not a replacement for formal support or bug reporting procedures through the XSEDE Help Desk. You must be logged in to post to the user forums.

« Back

Is an entire node on Forge the minimum allocation?

Combination View Flat View Tree View
Threads [ Previous | Next ]
Hi, after looking through the Forge users guide, I had one question:
When submitting a job, is the minimum resource allocation always an entire node (16 CPU cores + 6 CUDA devices)? (I see that the charging formula is in units of nodes, not CUDA devices.)

We typically run our software locally in a mode where a job only uses 4 CPU cores and 1 CUDA device, but if the Forge scheduler is reserving an entire node for a job, our code will tremendously underutilize the hardware. We could adjust things to use multiple CUDA devices per job without too much trouble, however. Should we go that direction?