AVX512 for PQFastScan #3276
Closed
AVX512 for PQFastScan #3276
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
|
Sorry for the late answer. |
|
@mdouze has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
1 similar comment
|
@mdouze has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
re-imported |
alexanderguzhva
force-pushed
the
avx512_pqfastscan
branch
from
74e4ae9
to
182affc
Compare
|
@mdouze got some problems with raft, rebased on top of master |
Signed-off-by: Alexandr Guzhva <[email protected]>
alexanderguzhva
force-pushed
the
avx512_pqfastscan
branch
from
182affc
to
515c0c6
Compare
|
@mdouze has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
abhinavdangeti
pushed a commit
to blevesearch/faiss
that referenced
this pull request
Jul 12, 2024
Summary: AVX-512 implementation for PQFastScan for QBS. For local benchmarks on 4th gen Xeon, the QPS is up to 10% higher, mostly for a single query case. But as far as I remember, production cases would show higher performance improvements. * Baseline `benchs/bench_ivf_fastscan_single_query.py` (sift1M): https://gist.github.com/alexanderguzhva/c9cde2cb5e9c7675f429623e6faa9fbf * Candidate `benchs/bench_ivf_fastscan_single_query.py` (sift1M): https://gist.github.com/alexanderguzhva/4e8530073a108f73771d38e55bc45b17 * Baseline `benchs/bench_ivf_fastscan.py` (sift1M): https://gist.github.com/alexanderguzhva/9eb03ed60354d7e76cfa25e676f983ac * Candidate `benchs/bench_ivf_fastscan.py` (sift1M): https://gist.github.com/alexanderguzhva/3cbfeba1364dd445a2bb52455966979e mdouze should I modify `pq4_fast_scan_search_1.cpp` as well? It is somewhat cumbersome to dig through various possible sub-implementations Pull Request resolved: facebookresearch#3276 Reviewed By: junjieqi Differential Revision: D54943632 Pulled By: mdouze fbshipit-source-id: 3d70066e9779039559b1734c2be99bf439058246
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.
AVX-512 implementation for PQFastScan for QBS.
For local benchmarks on 4th gen Xeon, the QPS is up to 10% higher, mostly for a single query case. But as far as I remember, production cases would show higher performance improvements.
benchs/bench_ivf_fastscan_single_query.py(sift1M): https://gist.github.com/alexanderguzhva/c9cde2cb5e9c7675f429623e6faa9fbfbenchs/bench_ivf_fastscan_single_query.py(sift1M): https://gist.github.com/alexanderguzhva/4e8530073a108f73771d38e55bc45b17benchs/bench_ivf_fastscan.py(sift1M): https://gist.github.com/alexanderguzhva/9eb03ed60354d7e76cfa25e676f983acbenchs/bench_ivf_fastscan.py(sift1M): https://gist.github.com/alexanderguzhva/3cbfeba1364dd445a2bb52455966979e@mdouze should I modify
pq4_fast_scan_search_1.cppas well? It is somewhat cumbersome to dig through various possible sub-implementations