Releases · tensorchord/pgvecto.rs

02 Feb 05:16

VoVAllen

v0.2.0

be5a816

v0.2.0

New features:

VBASE Mode Integration: pgvecto.rs integrates the cutting-edge VBASE method from OSDI 2023 and set it as default, substantially refining vector search efficiency within PostgreSQL. This enhancement not only accelerates the standard search process but also ensures accurate results, even when complex filters and join operations are applied.
FP16 Support: Users can now store their embeddings in PostgreSQL using half the float32 size, significantly improving latency. Our experiments indicate that this optimization has a negligible impact on final recall less than 1%.
Asynchronous Indexing: Insertion operations are now non-blocking, ensuring a smoother, more efficient data insertion and indexing process.
Doubled Query Performance: pgvecto.rs 0.2 boasts a query performance that is twice as fast as that of 0.1, marking a significant leap forward in our system's efficiency.
Enhanced Visibility into Internal Status: The new pg_vector_index_stat v offers users a transparent view into the indexing internals of pgvecto.rs. This feature allows for real-time monitoring of index construction, configuration adjustments, and detailed statistical analysis, fostering a more intuitive and controlled environment.

Breaking Changes:

Schema Modification for Installation: pgvecto.rs has shifted its installation location from the public schema to the vectors schema. This change necessitates updates in schema references within your queries and scripts.
Default Search Mode Update: The default search mode has been updated to VBASE. To revert to the basic search mode, users must manually execute the command SET vectors.search_mode = basic;. This change ensures that users benefit from the enhanced performance and accuracy of VBASE by default.
Revised prefilter Option: The semantics of the prefilter option have been redefined. The prefilter option now specifically controls whether deleted points are skipped from the search process. Meanwhile, the filter scenarios previously handled by this option are now adeptly managed by the VBASE mode, which offers superior accuracy and performance.
Index Compatibility and Reindexing Requirement: Indexes created with versions prior to 0.2 are not compatible with the 0.2 update. Users upgrading from earlier versions must reindex their data to ensure compatibility and to take full advantage of the new features and improvements introduced in pgvecto.rs 0.2.

Improvements

We've improved pgvecto.rs's reliability in this major release by adding more tests, improving error messages for easier troubleshooting, and refactoring the code for increased stability and smoother performance.

Assets 8

1 Join discussion

30 Jan 09:37

usamoi

v0.2.0-alpha.2

a237466

v0.2.0-alpha.2 Pre-release

Pre-release

Full Changelog: v0.2.0-alpha.1...v0.2.0-alpha.2

Assets 8

29 Jan 14:08

VoVAllen

v0.2.0-alpha.1

6332e1c

v0.2.0-alpha.1 Pre-release

Pre-release

Full Changelog: v0.2.0-alpha...v0.2.0-alpha.1

Assets 8

24 Jan 07:24

VoVAllen

v0.2.0-alpha

06e417c

v0.2.0-alpha Pre-release

Pre-release

What's Changed

ci: bump release version by @usamoi in #186
ci: fix release ci by @usamoi in #187
ci: fix nightly release timezone by @usamoi in #189
fix: do not config cross compilation in config.toml, document it in docs by @usamoi in #198
fix: fallback for memfd (kernel 3.17) by @usamoi in #204
feat: allow user to set vbase range by @silver-ymz in #206
chore: stable latest by @cutecutecat in #209
fix: upgrade instructions by @usamoi in #203
chore: move memfd detect to detect crate by @usamoi in #213
feat: add nprobe into guc by @cutecutecat in #212
feat: show index size in index info table by @kemingy in #222
chore: fine-grained upgrade hint by @usamoi in #220
chore: upgrade scripts for unsupported version by @usamoi in #221
chore: fix python action image tag by @kemingy in #229
chore: fix typos by @kemingy in #228
chore: friendly OOM message by @usamoi in #223
docs: development docs by @usamoi in #226
docs: add cutecutecat as a contributor for code by @allcontributors in #235
chore: Refine readme and replace docs with docs.pgvecto.rs by @gaocegege in #234
fix: typo by @usamoi in #237
fix: cosine distance by @kemingy in #231
fix: simpler logic for IPC connection reusing by @usamoi in #240
refactor: rework vbase by @usamoi in #238
chore: add crash test by @cutecutecat in #242
refactor: orthogonal support of vbase and prefilter by @usamoi in #248
chore: stable schema generation by @usamoi in #254
docs: advice developers to setup all PostgreSQL versions by @usamoi in #265
feat: Add vector to vecf16 CAST by @JoePassanante in #266
docs: add JoePassanante as a contributor for code by @allcontributors in #271
chore: add test for partition/partial/fp16 index by @cutecutecat in #276
docs: Refine readme by @VoVAllen in #274
chore: set schema and search_path by @usamoi in #273
fix: docker CMD instruction by @usamoi in #278
feat: transactional CREATE/DROP by @usamoi in #279
fix: promise flushing when committing by @usamoi in #280
chore: Add test for dropping index when indexing by @xieydd in #283
feat: add compat with pgvector by @cutecutecat in #256
chore: upload 0.2.0 schema install script by @usamoi in #289
fix: correct fp16 dot operation by @silver-ymz in #291
chore: upload --0.2.0 schema update script by @usamoi in #290
fix: prevent overflow at kmeans by @cutecutecat in #287
fix: checking of user options by @usamoi in #297
feat: support pgvector gucs compact by @cutecutecat in #296
chore: update the release & nightly action by @kemingy in #288
chore: fix release ci gh upload tag parameter by @kemingy in #300
chore: allow tag to have a v-prefix by @kemingy in #301
fix: check if prefilter enabled by @usamoi in #302
chore: nightly token that can trigger the release workflow by @kemingy in #303