Hot take PR: Use vlsvrs to read in compressed VDFs#430
Merged
alhom merged 27 commits intofmihpc:devfrom Mar 24, 2026
Merged
Conversation
…terix_vlsvrs_vdfs
Contributor
Author
|
Apparently the test image CI test is supposed to fail right now so the PR looks good |
Contributor
Author
|
Do we close this now in favor of banalysator? |
Member
We can just point the dependency to that one instead! |
…wheels), instructions and an optional dependency for building the backend locally (requires rust)
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
So this PR is a second attempt at reading in compressed VDFs produce by Asterix. In this attempt vlsvrs is used when VDFs are compressed. By default the supported methods
are :
UPDATE:
At this point vlsvrs will build tinyAI and the MLP compressor (targeting CPU inference only) when either
nvccorhipccare on the system. In any other case it will skip compiling the MLP.(MLP is actually being phased out so support for that will be limited)
So in this case we have no python interfacing with any C/C++/CUDA libs since it is all done by vlsvrs.