Commit Graph

26 Commits

Author SHA1 Message Date
Andrey Vasnetsov
4437edb775 Weighted rrf (#8063)
* weighted rrf implementation

* test

* fmt

* fix edge

* validate number of sources and number of weights

* do not partial match

* upd schema

* review fixes

* update formula

* remove calcualtions from tests

* update comment, because AI have OCD

* fmt
2026-02-06 19:36:27 +01:00
Andrey Vasnetsov
921b1ce83d Prevent unoptimized updates (#7643)
* wip: wait for optimization before applying update

* implement api parameter

* nits

* fix deadloack when no optimizers are running

* release handle mutex

* Derive PartialEq

* Remove unused function

* fix missing kb conversion in threshold

* Update lib/api/src/grpc/proto/collections.proto

Co-authored-by: Tim Visée <tim+github@visee.me>

* sync comment change

---------

Co-authored-by: Arnaud Gourlay <arnaud.gourlay@gmail.com>
Co-authored-by: timvisee <tim@visee.me>
Co-authored-by: Tim Visée <tim+github@visee.me>
2026-01-14 16:37:03 +01:00
Daniel Boros
1cbf297f9e feat/shard-ops-timeout (#7750)
* feat: add shard-ops timeout

* feat: add shard-ops timeout

* feat: add missing timeout

* fix: format

* fix: type serialization

* fix: bad wal flush logic

* feat: add seperate wait_time

* fix: linter issues

* fix: linter

* chore: make select more readable

* fix: linter issues

* chore: remove comment

* chore: adjust comment

* chore: update definition

* chore: remove extra wait_timeout -> make timeout global

* chore: remove wait_timeout from point_ops

* chore: remove wait_timeout from grpc

* refactor: make shard_ops more readable

* chore: remove timeout from update_all

* feat: propagate wait_timeout error

* fix: clippy recommendation

* chore: update wait timeout

* refactor: update merge_successful_update_results

* fix: tests inconsistency

* fix: revert global ord, partialord for UpdateStatus

* fix: integration test

* fix: res ingoring

* fix: review requests

* fix: clippy

* chore: revert back futures unordered

* fix: resolve response status

* force timeout for remote operations

* fix: pr reviews

* fix: switch to duration

* Bump OpenAPI spec

* feat: add serde_as

* fix: timeout statu condition

* Minor tweaks

* Use tokio timeout directly rather than deadline

* Revert schema generator permissions

* Remove obsolete semi-colon

---------

Co-authored-by: generall <andrey@vasnetsov.com>
Co-authored-by: timvisee <tim@visee.me>
2026-01-09 10:24:45 +01:00
Andrey Vasnetsov
37b0f71e26 Move Query Context into shard crate (#7459)
* Move modifier to segment level

* write idf modifier into segment config

* fmt

* use proper query context in edge

* simplify init_query_context
2025-10-29 11:39:16 +01:00
Andrey Vasnetsov
29a1197c4e derive Eq for Filter (and underlying types) (#7419)
* derive Eq for Filter (and underlying types)

* fix tests

* use ptr to compare and hash CustomIdChecker

* fix gpu test

* fmt

* use OrderedFloat directly

* post-rebase fixes

* fmt

---------

Co-authored-by: Luis Cossío <luis.cossio@outlook.com>
2025-10-20 17:27:03 +02:00
Ivan Boldyrev
ad31c428e3 Fix Tokio race condition in benchmarks (#7311)
Shutdown `shard` explicitly at the end of benchmark to avoid error

```
A Tokio 1.x context was found, but it is being shutdown.
```

Co-authored-by: Ivan Boldyrev <ivan.boldyrev@qdrant.com>
2025-09-25 16:06:19 +02:00
Andrey Vasnetsov
e744bd6014 implement custom collection metadata (#7123)
* implement custom collection metadata

* persist metadata change

* Also unset a key in the test

---------

Co-authored-by: timvisee <tim@visee.me>
2025-08-26 11:49:17 +02:00
Luis Cossío
a2d3270eb8 Custom RRF k parameter (#7065)
* allow custom K parameter for RRF

* generate grpc docs and openapi

* use tagged type approach for parametrized fusions

* use params approach in grpc

* simplify api structure

* upd schema

* nits

* rest: parameterized rrf as query variant

* consistency in doc comments

---------

Co-authored-by: Andrey Vasnetsov <andrey@vasnetsov.com>
2025-08-20 15:05:27 -04:00
Roman Titov
0bfa8f61d5 Move SaveOnDisk into common crate (#7105) 2025-08-20 17:06:47 +02:00
Jai
57ae0dd044 Fix criterion batch_query_bench (#6825) (#7077) 2025-08-18 15:16:30 +02:00
Kumar Shivendu
34e18eb22c Allow retaining more closed segments (#6976)
* Retain more closed segments

* Add back rocksdb

* Update WalOptions/WalConfig across the code

* Use NonZeroUsize

* Expose via APIs

* Update gRPC docs

* Fix stoarge compat test

* recompile openapi.json with rocksdb

* default wal retain closed fn

* update openapi.json

* Use qdrant/wal latest commit and remove from config.yaml
2025-08-07 13:44:26 +02:00
Tim Visée
a5869beb99 Deprecations for Qdrant 1.15.0 (#6892)
* Deprecate init_from

* Mark memmap_threshold as deprecated

* Mark locks API as deprecated

* Mark RBAC collection access payload filter as deprecated

* Allow deprecations in tests and benchmarks
2025-07-17 16:10:45 +02:00
Tim Visée
3e536347e1 Bump Rust edition to 2024 (#6042)
* Bump Rust edition to 2024

* gen is a reserved keyword now

* Remove ref mut on references

* Mark extern C as unsafe

* Wrap unsafe function bodies in unsafe block

* Geo hash implements Copy, don't reference but pass by value instead

* Replace secluded self import with parent

* Update execute_cluster_read_operation with new match semantics

* Fix lifetime issue

* Replace map_or with is_none_or

* set_var is unsafe now

* Reformat
2025-02-25 11:21:25 +01:00
Andrey Vasnetsov
f7d0814ab6 IO resource usage permit (#6015)
* rename cpu_budget -> resource_budget

* clippy

* add io budget to resources

* fmt

* move budget structures into a separate file

* add extend permit function

* dont extend existing permit

* switch from IO to CPU permit

* do not release resource before aquiring an extension

* fmt

* Review remarks

* Improve resource permit number assertion

* Make resource permit replace_with only acquire extra needed permits

* Remove obsolete drop implementation

* allocate IO budget same as CPU

* review fixes

---------

Co-authored-by: timvisee <tim@visee.me>
2025-02-20 09:05:00 +01:00
Luis Cossío
af74d1b96a bump and migrate to rand 0.9.0 (#5892)
* bump and migrate to rand 0.9.0

also bump rand_distr to 0.5.0 to match it

* Migrate AVX2 and SSE implementations

* Remove unused thread_rng placeholders

* More random migrations

* Migrate GPU tests

* bump seed

---------

Co-authored-by: timvisee <tim@visee.me>
Co-authored-by: Arnaud Gourlay <arnaud.gourlay@gmail.com>
2025-01-28 16:19:11 +01:00
Jojii
a57af177ea Propagate hardware counter for more functions (#5844)
* Propagate hardware counter for more functions

* Minor improvements

* use vector_query_contexts hardware_counter
2025-01-24 16:33:44 +01:00
Jojii
e0d0e233d8 Timeout aware hardware counter (#5555)
* Make hardware counting timeout aware

* improve test

* rebuild everything

* fmt

* post-rebase fixes

* upd tests

* fix tests

---------

Co-authored-by: generall <andrey@vasnetsov.com>
2024-12-10 12:12:36 +01:00
Jojii
b83ace82f0 Per collection hardware measurements (#5453)
* Add HwMeasurementCollector

* Add hardware reporting to TOC + RequestHwCounter

* Pass HwMeasurementAcc by reference + Update accumulation

* Update tests and benchmarks

* update REST API

* Update gRPC API

* codespell

* Adjust internal API

* improve docs

* introduce drain to the HwMeasurementAcc

* fmt

* use drain to report to the collection counter

* implement hw metrics drain for internal and external queries

* fix drinage

* refactor rest models: move away from grpc crate

* fmt

* implement usage reporting to collection acc for rest api

---------

Co-authored-by: generall <andrey@vasnetsov.com>
2024-11-17 12:46:01 +01:00
Roman Titov
7e7348613e Add UUID to collection config (#5378)
* Add UUID to collection...

...and recreate collection, when applying Raft snapshot, if UUID of collection is different

* fixup! Add UUID to collection...

Remove UUID field from gRPC and exclude it from OpenAPI spec 🤡

* fixup! fixup! Add UUID to collection...

Always generate collection UUID 🤦‍♀️

* Raft snapshot recreate collection no expose UUID (#5452)

* separate colleciton config structure from API

* fmt

* Update lib/collection/src/operations/types.rs

Co-authored-by: Tim Visée <tim+github@visee.me>

---------

Co-authored-by: Tim Visée <tim+github@visee.me>

---------

Co-authored-by: Andrey Vasnetsov <andrey@vasnetsov.com>
Co-authored-by: Tim Visée <tim+github@visee.me>
2024-11-15 19:03:50 +01:00
Tim Visée
30fcbc298c Revert experiment: ignore WAL clocks in partial state (#5353)
* Revert "Experiment: in stream records, set cutoff to latest clocks receiver is guaranteed to have (#5375)"

This reverts commit e843647c95.

* Revert "Set replica state function does not need to be async anymore (#5379)"

This reverts commit fdf08e0a59.

* Revert "Experiment: disable clocks in initializing state, propagate ignore flag (#5372)"

This reverts commit 0d98f6fe1a.

* Revert "Experiment: also don't write clock tags to WAL in partial state (#5352)"

This reverts commit f69ec61848.

* Revert "Experiment: ignore clock tags when replica is in partial state (#5349)"

This reverts commit 1b8a38f70b.

* Keep change to not send updates to replicas in recovery state

Co-authored-by: Andrey Vasnetsov <andrey@vasnetsov.com>

---------

Co-authored-by: Andrey Vasnetsov <andrey@vasnetsov.com>
2024-11-08 14:48:53 +01:00
Tim Visée
0d98f6fe1a Experiment: disable clocks in initializing state, propagate ignore flag (#5372)
* Propagate flag for ignoring clocks on local shard from replica set

* Also ignore local clocks in initializing state, is similar to partial

* Remove previous logic for disabling clocks

* We can make static replica set state functions inlined const

* Fix typo
2024-11-05 17:46:02 +01:00
Jojii
73025ba9d1 Populate hardware counter to REST API (#5308)
* populate hardware counter

* make consume semantic explicit

* Merge pull request #5328

* add hardware info to more endpoints

---------

Co-authored-by: generall <andrey@vasnetsov.com>
2024-10-29 22:15:37 +01:00
Andrey Vasnetsov
e0d507dac7 Inference interface in REST and gRPC (#5165)
* include document & image objects into grpc API

* introduce image and object to rest api

* minor refactoring

* rename Vector -> VectorInternal

* decompose vector data structures

* add schema

* fmt

* grpc docs

* fix conversion

* fix clippy

* fix another conversion

* rename VectorInput -> VectorInputInternal

* replace grpc TryFrom with async functions

* fmt

* replace rest TryFrom with async functions

* add image and object into query rest

* separate inference related conversions

* move json-related conversions into a separate file

* move vector-related transformations into a separate file

* move more vector related-conversions into dedicated module
2024-10-09 10:15:46 +02:00
Jojii
e9bad464ff [Strict-Mode] Basic implementation (#4887)
* add CollectionRequestVerification

* add to api

* rebase

* improve implementation

* implement strict mode for SearchRequest+Batch

* improve code + fix Clippy

* improve error handling

* restructure StrictModeVerification trait

* generate docs

* check `enabled` option

* review remarks

* rename StrictModeConfigDiff in grpc

* use missing payload detection from issue api

* performance improvement

* decouple extractor from issues (#4945)

* some review remarks

* don't default to empty functions in StrictModeVerification trait

* update openapi

* filter_limit => query_limit

* replace discovery_max_context_size and recommend_max_examples with max_input_examples

* review remarks

* review fix: include possible index types into error message

* review remarks

---------

Co-authored-by: Luis Cossío <luis.cossio@qdrant.com>
Co-authored-by: generall <andrey@vasnetsov.com>
2024-08-29 10:49:23 +02:00
Arnaud Gourlay
196f889a5a Non blocking retrieve with timeout and cancellation support (#4844)
* Non blocking retrieve with timeout and cancellation support

* apply timeout for extra retrieve in rescoring
2024-08-08 12:41:33 +02:00
Luis Cossío
6461e06a7f perf: Use Mutex in PrefetchResults (#4664)
* add bench

* add rrf bench

* add rescore bench

* use RwLock instead of Cow

* lower limits in bench

* fix after rebase

* use sync mutex

* fmt

---------

Co-authored-by: generall <andrey@vasnetsov.com>
2024-07-15 10:13:48 +02:00