Add Similarity Search optimization guides - #20
Merged
Conversation
- Remove BIOS and hardware recommendation sections - Add high-level LVQ and LeanVec description - Replace benchmarks with data from Redis tech dive blog - Mention SVS can be used directly, integrations in progress - Remove contributing section - Focus on compression selection and performance data
- Add FAISS optimization guide with SVS index types - Cover IndexSVSVamana, IndexSVSVamanaLVQ, IndexSVSVamanaLeanVec - Include factory string format and examples - Add compression selection guide - Include benchmark data from Intel SVS - Update overview and main README
- Remove benchmark section (no official FAISS SVS performance data) - Update overview to remove specific performance claims - Clarify Intel-only requirement in FAQ - Keep technical how-to guidance
- Add licensing warning for LVQ/LeanVec (AGPLv3/SSPLv1 incompatible) - Add language spec to FAISS factory string code block
ahuber21
suggested changes
Feb 10, 2026
ahuber21
reviewed
Feb 11, 2026
ahuber21
left a comment
ahuber21
left a comment
There was a problem hiding this comment.
Dropping my change request to not block anything while I'm OOO. Still would like to see the FAISS doc sorted out though.
ahuber21
approved these changes
Feb 11, 2026
Collaborator
|
@napetrov Do you plan to drop FAISS document in this PR? |
Contributor
Author
yes, removed. Would give opportunity to look on this pr for couple other people |
mihaic
reviewed
Feb 12, 2026
- Fix SVS intro: 'is integrated' instead of 'working on integrating' - Clarify Level 2 compression: LVQ encodes residuals, LeanVec encodes full dimensionality - Fix LeanVec4x8 description with reduced/full dimensionality detail - Fix DISTANCE_METRIC guidance: L2 for normalized embeddings - Add 'slower build' note to CONSTRUCTION_WINDOW_SIZE - Fix compression ratios: LVQ4x8 ~2.5x, LeanVec notation with f factor - Add LeanVec dimensionality reduction factor explanation - Remove bash markers from Redis command blocks - Fix SVS-VAMANA parameter count (14 → 12) in LeanVec example - Remove EPSILON row (not SVS-specific, range search only) - Fix attribution: Redis benchmarking (authors are Redis) - Fix precision metric: remove '+' (calibrated value) - Clarify SVS-VAMANA effectiveness: 'improving throughput' Co-authored-by: mihaic <mihaic@users.noreply.github.com>
mihaic
suggested changes
Feb 17, 2026
mihaic
left a comment
mihaic
left a comment
Member
There was a problem hiding this comment.
A few of my earlier suggestions are unaddressed and there is still a reference to the deleted Faiss document.
Co-authored-by: Mihai Capotă <mihai@mihaic.ro>
- Remove 'bash' code fence marker from FT.SEARCH example - Remove SEARCH_BUFFER_CAPACITY row (not SVS-specific, range search only)
mihaic
suggested changes
Feb 25, 2026
mihaic
left a comment
mihaic
left a comment
Member
There was a problem hiding this comment.
I marked some previous comments as still relevant. Please have a look.
… Redis OS link, reorder FAQ steps
Contributor
Author
|
Addressed remaining review comments in commit 82c959e:
|
mihaic
suggested changes
Feb 26, 2026
Contributor
Author
|
Done. Added REDUCE back to the table in commit 0ab8dbb. |
mihaic
approved these changes
Feb 27, 2026
aguerreb
reviewed
Mar 4, 2026
aguerreb
left a comment
aguerreb
left a comment
There was a problem hiding this comment.
Looks good overall, made a few comments.
- Update intro: add LVQ/LeanVec mention in Redis intro sentence - Add Vamana algorithm reference (Subramanya et al., NeurIPS 2019) - GRAPH_MAX_DEGREE: add note 'Equivalent to HNSW M x 2' - Remove upper bound 3072: change to '768+ dimensions' - Benchmark LVQ8 note: clarify LeanVec recommended for Cohere/DBpedia in production - Ingestion trade-offs: replace 'compression overhead' with broader description - FAQ recall steps: swap steps 2 and 3 (REDUCE before TRAINING_THRESHOLD) - FAQ ARM: clarify HNSW preferable due to both faster search and ingestion
aguerreb
approved these changes
Mar 4, 2026
Contributor
Author
|
@jasukhar - i think we ready to merge this |
harp-intel
approved these changes
Mar 6, 2026
rsiyer-intel
approved these changes
Mar 6, 2026
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
6 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.
Summary
This PR adds documentation for optimizing vector similarity search workloads on Intel Xeon processors.
New Content
software/similarity-search/README.md— Overview of Intel's compression technologies (LVQ, LeanVec) for vector searchsoftware/similarity-search/redis/README.md— Complete guide for Redis SVS-VAMANA configuration and tuningKey Topics Covered
Benchmarks Highlighted
Based on Redis and Intel benchmarking:
Related: Intel Scalable Vector Search library — https://intel.github.io/ScalableVectorSearch/