Addition of node type details and cleaning up shared vs. exclusive info #339
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
vanderwb
approved these changes
Dec 24, 2025
Collaborator
vanderwb
left a comment
vanderwb
left a comment
There was a problem hiding this comment.
Sorry it took so long to get a review in here. I agree those sections are a bit basic, and I can't think of much to add at the moment, but I think we should keep them and can expound as we think of useful guidance to add.
Thanks for fixing this up!
| The L40 nodes are also capable of basic GPGPU tasks like AI inference and are less utilized than the nodes within the GPGPU queue. This could significantly reduce your wait time in the queue. | ||
|
|
||
| !!! info | ||
| Data and Visualization node type requests will be submitted to the `vis` queue and these GPUs are shared among multiple users on a node. The maximum selectable GPUs is one and exclusive access cannot be guaranteed on the Data and Visualization node types. If you need exclusive node access, use the ML & GPGPU node types. |
Collaborator
There was a problem hiding this comment.
I would word this as "the maximum selectable ngpus is one"
Instead of GPUs, use the resource type to be clear that the number of gpus in the select statement is a max of 1.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Deleting the sections for "Shared vs. Exclusive" and distilled them into an example with a tip on how these impact performance or queue times.
Added sections for each node type. The HTC and GPGPU sections feel a bit basic so I'm open to feedback either to add detail or just remove the sections. The largemem and vis sections have a few tricks that are worth keeping.