fix: GPU interpretation of RNTuple data working and implement nvCOMP decompression with nvCOMP library #1557
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Outdated suggestions cannot be applied.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.
This PR brings GPU interpretation of RNTuple data back in working order and transitions GPU decompression to using the nvCOMP python package. This addresses issues #1514 and #1497.
After some updates to post-processing and deserialization of RNTuple data, GPU interpretation no longer worked. A
cupy.ndarraycannot be post processed in place as was done. Thereforepost_processwas renamed topost_process_in_placefor use onnumpy.ndarray, and a newpost_processfunction which can perform a similar function oncupy.ndarrayis added. Further minor tweaks to the GPU interpretation made to match behavior of CPU.Kvikio deprecated their bindings to nvCOMP in this PR rapidsai/kvikio#714. GPU decompression is now handled with the official Nvidia nvCOMP python package.