Add pooling mode to device bench #1194
Closed
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
✅ Deploy Preview for eclectic-stroopwafel-199537 ready!
To edit notification comments on pull requests, go to your Netlify site settings. |
facebook-github-bot
pushed a commit
that referenced
this pull request
Jul 13, 2022
Summary: This is an attempt to optimize the nobag forward kernel for tables whose embedding dim is smaller or equal to 32. I was exploring this as some of our production models have embedding_dim = 32. The optimization results in 10%~30% enhancement for small embedding_dim and could be applied to other kernels. However, it's worth noticing that a 10% enhancement on 1 kernel can barely have any effect on the overall training speed. Therefore, I'm totally fine with whether this optimization gets accepted, just trying to share some ideas we had to prevent others' repetitive work :) The main rationale is that the current implementation will use all 32 threads in a warp to load 1 embedding vector, which means when the embedding dim is smaller than 128, some threads in the warp do nothing but wait. This PR will split threads into groups, e.g. for embedding_dim=32, each group has 8 threads, and let the threads process each embedding vector in group instead of in warp. The performance enhancement of this trick is benchmarked with #1194: ``` python bench/split_table_batched_embeddings_benchmark.py device --pooling=none --iters=1000 --embedding-dim=$EMB_DIM ``` And figures are: |embedding dim|4|8|16|32| |---------------|-|-|-|-| |before(us)|136|138|142|154| |after(us)|96|100|113|141| Thank you for your time on this PR and it will be great if you could share your thoughts on this type of optimization :) Pull Request resolved: #1197 Reviewed By: jasonjk-park Differential Revision: D37739239 Pulled By: jianyuh fbshipit-source-id: 23b35d74eed28f977793cb52e311ba0f824ac634
|
@colin2328 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
Thanks! Just curious in your use case, do you have both unpooled embedding (none) and pooled embedding (sum), or only unpooled embedding (none)? |
jianyuh
reviewed
Jul 14, 2022
| @@ -191,6 +205,7 @@ def device( # noqa C901 | |||
| weights_precision=weights_precision, | |||
| stochastic_rounding=stoc, | |||
| output_dtype=output_dtype, | |||
| pooling_mode=pooling_mode, | |||
There was a problem hiding this comment.
Could you update the BW calculation formula in Line 254 and Line 279 ? Basically referring to
FBGEMM/fbgemm_gpu/bench/split_table_batched_embeddings_benchmark.py
Lines 983 to 991 in b3e129c
jianyuh
approved these changes
Jul 15, 2022
|
@colin2328 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
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.
This would help us benchmark EmbeddingCollection in torchrec.
Thank you for your time in reviewing this PR :)