[Save this search]

Status
All
   Fixed (7187)
  Closed (4439)
✓Open (2528)
   Won't Fix (543)
   Duplicate (297)
   Invalid (217)
   Not A Problem (195)
Issue type
✓All
  Issue (2229)
  PR (299)
Author relation
✓All
  None (2031)
  Member (594)
  Contributor (409)
  New contributor (56)
Created
✓All
  Past day (9)
  Past 2 days (11)
  Past 3 days (14)
  Past week (16)
  Past month (51)
  Past 3 months (82)
  Past 6 months (156)
  Past year (282)
Updated
✓All
  Past day (15)
  Past 2 days (22)
  Past 3 days (26)
  Past week (36)
  Past month (81)
  Past 3 months (118)
  Past 6 months (399)
  Past year (506)
Updated ago
✓All
  > 1 day ago (2513)
  > 2 days ago (2506)
  > 3 days ago (2502)
  > 1 week ago (2492)
  > 1 month ago (2447)
  > 3 months ago (2410)
  > 1 year ago (2021)
Comment count
✓All
  0 (556)
  1 (356)
  2 - 5 (792)
  6 - 10 (425)
  10 - 20 (305)
  > 20 (162)
Reaction count
✓All
  0 (2248)
  1 (167)
  2 - 5 (91)
  6 - 10 (16)
  10 - 20 (5)
  > 20 (1)
Review Requested
✓All
  jpountz (14)
  mikemccand (8)
  benwtrent (5)
  iverase (4)
  romseygeek (4)
  rmuir (3)
  dweiss (3)

See all 23...
Mentioned
✓All
  jpountz (90)
  mikemccand (71)
  rmuir (49)
  benwtrent (43)
  uschindler (38)
  msokolov (37)
  vigyasharma (18)

See all 194...
Reviewed
✓All
  jpountz (23)
  mikemccand (22)
  rmuir (13)
  uschindler (12)
  benwtrent (12)
  dweiss (11)
  msokolov (10)

See all 57...
Commented
✓All
  asfimport (1369)
  github-actions[bot] (226)
  jpountz (140)
  mikemccand (117)
  rmuir (95)
  benwtrent (81)
  msokolov (75)

See all 284...
User
✓All
  asfimport (1811)
  github-actions[bot] (268)
  mikemccand (201)
  jpountz (199)
  rmuir (127)
  benwtrent (119)
  msokolov (94)

See all 432...
Last comment user
✓All
  asfimport (1326)
  github-actions[bot] (221)
  jpountz (32)
  mikemccand (30)
  rmuir (23)
  benwtrent (20)
  gsmiller (15)

See all 153...
Draft
✓All
  No (248)
  Yes (51)
Component
✓All
  core (590)
  analysis (150)
  highlighter (47)
  spatial (41)
  facet (40)
  queryparser (27)
  test-framework (26)

See all 23...
Type
✓All
  enhancement (1160)
  bug (728)
  task (197)
  test (77)
  documentation (22)
Labels
✓All
  Stale (216)
  legacy-jira-fix-versio... (214)
  legacy-jira-fix-versio... (169)
  affects-version:4.0-ALPHA (81)
  tool:build (54)
  vector-based-search (47)
  affects-version:6.0 (37)

See all 156...
Commits?
✓All
  No (2528)
Reporter
✓All
  rmuir (271)
  mikemccand (153)
  jpountz (126)
  dsmiley (68)
  uschindler (52)
  iverase (42)
  dweiss (40)

See all 737...
Assignee
✓All
  Unassigned (2288)
  mikemccand (35)
  uschindler (31)
  romseygeek (27)
  dsmiley (22)
  rmuir (14)
  jpountz (11)

See all 44...
  
Filters: Status (Open),  Issue type,  Author relation,  Created,  Updated,  Updated ago,  Comment count,  Reaction count,  Review Requested,  Mentioned,  Reviewed,  Commented,  User,  Last comment user,  Draft,  Component,  Type,  Labels,  Commits?,  Reporter,  Assignee

#15676 PR: Feature/collaborative hnsw search
17.6 minutes ago  12 comments  0 votes  0 watches  benwtrent, github-actions[bot], krickert, navneet1v
[HNSW] Collaborative Search via Dynamic Threshold Feedback > **Status: Experimental** - This ... Full distributed multi-shard testing will happen in a follow-up PR for OpenSearch.
    benwtrent 36.1 minutes ago:  I don't understand your graphs. I would expect the following results: - Single shard to have the lowest vector ops and the ... Your graphs show there is zero benefit for collaborative and that splitting data across shards ...
    krickert 51 minutes ago:  Ok.. I've completed the local Pareto sweep (single-shard baseline, multi-shard independent, multi-shard ... Independent fork matches upstream baseline recall exactly, and collaborative is generally ...

#15730: MemorySegmentIndexInput absolute read methods might modify position
42.7 minutes ago  0 comments  0 votes  0 watches  viliam-durina
Description The absolute read methods in RandomAccessInput should read from the given position, ... However MemorySegmentIndexInput's absolute read methods modify the input's position if the read ... The following test reproduces it: `java public class TestMemorySegmentIndexInput extends ...

#15734: Explore more granular vector quantization?
46.3 minutes ago  2 comments  0 votes  0 watches  kaivalnp
Description Today, Lucene supports 8, 4, 2 and 1 bit quantization. Each quantization level typically has an upper bound of recall with exact KNN that it can produce ( ... Any algorithm operating on quantized scores _alone_ cannot go beyond this recall (e.g. tweaking ...
    kaivalnp 46.3 minutes ago:  This "expansion" to a higher dimension is probably better to do outside Lucene (ideally the model ... , but sharing because I found it interesting that it works on existing vectors too! Perhaps we can add an option to luceneutil to try non-standard quantization options?
    kaivalnp 56.7 minutes ago:  However, I understand that packing vectors with arbitrary quantization levels + supporting various ... (e.g. represent 1024 dimension vectors in 1536 dimensions, then quantize to 4 bits: we get the ... We need a rotation because quantization occurs per-dimension, so any dimension with all zeros is ...

#15697: For 4-bit quantized vectors, should we change the scheme of unpacking nibbles?
1.6 hours ago  3 comments  0 votes  0 watches  benwtrent, kaivalnp, mccullocht, shubhamvishu
Description 4-bit quantized vectors are stored as nibbles, two packed in a single byte -- where ... Example: [1 5][2 6][3 7][4 8] For performing individual vector computations, we "unpack" the ... For higher dimension vectors, can this lead to CPU cache line misses?
    mccullocht 1.6 hours ago:  The "unpacked" vector is the query vector and is likely to be in the L1 cache for the duration of ... A load from the L1 cache should be a speed bump -- a cycle or two at most. The packed vector is where you are most likely to miss on caches and we read that one linearly, ...
    kaivalnp 8 days ago:  Also pasting a sample JMH benchmark (from 15698) run using: ` java --module-path lucene/ ... *SinglePacked.*" -p size=1024 ` ` Benchmark ... Individual JMH iterations started strong, but performance dropped after some iterations.

#15729 PR: Doc values skipper: Disable prefetching
2.5 hours ago  1 comments  0 votes  0 watches  github-actions[bot], rmuir, spinscale
This change requires some explanation. While trying to understand how doc value skippers work, I noticed on skipper generation ... While for most other usage in the doc values producer this seems correct, in the case of the doc ...
    rmuir 2.5 hours ago:  Yeah I don't think we should prefetch pages until the first sequential read. Should not happen inside random access functions such as advance() or similar. This is just my opinion.

#15722 PR: Add the ability to do prefetch and do prefetch during vector search
3.5 hours ago  13 comments  0 votes  0 watches  benwtrent, github-actions[bot], mccullocht, mikemccand, navneet1v, vigyasharma
Description This change add the ability to do prefetch during Approximate search and exact search ... The changes include: 1. New prefetch interface on KNNVectorValues and RandomVectorScorer. 2.
    navneet1v 3.5 hours ago:  > We need to benchmark with quantized vectors, where the vector ops during traversal are magnitudes ... I am working on the benchmarks.
    benwtrent 6.5 hours ago:  I don't mind the interface. But it does seem that the "scorer.bulkScore" should "just do the right thing" and prefetch if ... I think this will harm the "preferred path", which is the one where all vectors are in MMAP'd space ...

#15731: Tessellation error for valid MultiPolygon
4.3 hours ago  1 comments  0 votes  0 watches  iverase, roblburris
Description Similar to https://github.com/apache/lucene/issues/15695, we have another geometry ... Possible malformed shape detected. error when calling Tessellatortessellate: ` java.lang.IllegalArg ... Possible malformed shape detected.
    iverase 4.3 hours ago:  This is a different issue as the ones before. If we look how holes are getting connected to the outer ring, there is one hole that not doing the ... <img width="500" height="740" alt="Image" src="https://github.com/user-attachments/assets/ ...

[21.4 msec search, 22.4 msec total]