[FEA] Buffered and In-Memory Sampling #4628
Open
Conversation
This file contains 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
alexbarghi-nv
added
feature request
alexbarghi-nv
added
breaking
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Adds support for buffered sampling, and supports in-memory sampling as a form of buffered sampling. Also separates the sample readers and sample writers into their own files to make the code easier to read.
Adds support for buffered in-memory sampling in cuGraph-PyG. Updates the PyG examples to add in-memory sampling as an option.
Also fixes a bug where COO output of the distributed sampler (which is rarely used) was overwriting previous batches.
Based on benchmarks on H100, buffered in-memory sampling is 3.2x faster than disk sampling. This results in an overall speedup of 1.25x. This speedup is consistent on SG, MG, and MNMG up to 16 GPUs for ogb-papers100M.
Closes rapidsai/cugraph-gnn#14