zhagnlu
32f575be0f
enhance: change bitmap index mmap mode to view mode ( #38179 )
...
#38138
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2024-12-08 17:26:41 +08:00
Xianhui Lin
6d0a4fdb31
fix: Fix bug for Search fails with filter expression contains underscore ( #38085 )
...
Enhance the matching for elements within the UnaryRangeArray
https://github.com/milvus-io/milvus/issues/38068
---------
Signed-off-by: Xianhui.Lin <xianhui.lin@zilliz.com>
2024-12-05 10:18:40 +08:00
tinswzy
262f6db3d8
enhance: Add mmap file usage metric ( #38193 )
...
issue: #38156 Add mmap file usage metric
Signed-off-by: tinswzy <zhenyuan.wei@zilliz.com>
2024-12-04 16:12:47 +08:00
aoiasd
87aa9a0f2d
fix: empty analyzer params not use standard tokenizer ( #38148 )
...
relate: https://github.com/milvus-io/milvus/issues/35853
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-12-04 14:58:39 +08:00
sre-ci-robot
3445b8bd44
[automated] Update Knowhere Commit ( #38192 )
...
Update Knowhere Commit
Signed-off-by: sre-ci-robot sre-ci-robot@users.noreply.github.com
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2024-12-04 02:30:39 +08:00
congqixia
767b7e6218
enhance: Use fdopen, fwrite to reduce direct syscall ( #38157 )
...
`File.Write` and `File.WriteInt` use `write`, which may be just direct
syscall in some systems. When mappding field data and write line by
line, this could cost lost of CPU time when the row number is large.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-12-03 15:24:39 +08:00
Bingyi Sun
90064cd47b
fix: Fix variable redeclaration in term filter ( #38045 )
...
https://github.com/milvus-io/milvus/issues/38046
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-12-02 15:10:38 +08:00
Zhen Ye
c6dcef7b84
enhance: move segcore codes of segment into one package ( #37722 )
...
issue: #33285
- move most cgo opeartions related to search/query into segcore package
for reusing for streamingnode.
- add go unittest for segcore operations.
Signed-off-by: chyezh <chyezh@outlook.com>
2024-11-29 10:22:36 +08:00
sre-ci-robot
0894ed0016
[automated] Update Knowhere Commit ( #38082 )
...
Update Knowhere Commit
Signed-off-by: sre-ci-robot sre-ci-robot@users.noreply.github.com
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2024-11-29 01:08:36 +08:00
Bingyi Sun
e6af806a0d
enhance: optimize self defined rust error ( #37975 )
...
Prepare for issue: https://github.com/milvus-io/milvus/issues/37930
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-28 20:30:36 +08:00
congqixia
cb6542339e
enhance: Mark cgo thread with tag name ( #38000 )
...
Related to #37999
This PR add `SetThreadName` API for marking cgo thread and utilize it
when initializing cgo worker.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-11-26 11:22:35 +08:00
Zhen Ye
fbb68ca370
enhance: make all index operation async scheduled by tokio ( #37946 )
...
issue: #37851
related pr: https://github.com/milvus-io/tantivy/pull/3
Signed-off-by: chyezh <chyezh@outlook.com>
2024-11-25 10:12:34 +08:00
sre-ci-robot
ed73dfca3f
[automated] Update Knowhere Commit ( #37965 )
...
Update Knowhere Commit
Signed-off-by: sre-ci-robot sre-ci-robot@users.noreply.github.com
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2024-11-25 01:28:32 +08:00
zhagnlu
62af24c1a1
fix: change search latency metric from us unit to ms unit ( #37806 )
...
#37805
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2024-11-24 17:26:33 +08:00
Bingyi Sun
700a448a54
fix: Escape prefix before search in inverted index ( #37925 )
...
issue: https://github.com/milvus-io/milvus/issues/37912
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-22 14:10:33 +08:00
Bingyi Sun
06d73cf2e2
enhance: Remove raw tokenizer register. ( #37886 )
...
tantivy already register raw tokenizer by default
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-22 12:02:32 +08:00
Zhen Ye
1dc1a97e65
fix: use different thread pool for scheduler and merger ( #37911 )
...
issue: #37895
related pr: https://github.com/milvus-io/tantivy/pull/2
Signed-off-by: chyezh <chyezh@outlook.com>
2024-11-21 21:34:33 +08:00
Zhen Ye
f3a36f8a29
fix: use global pool but not dedicated pool for every index ( #37852 )
...
issue: #37851
- make a global thread pool at tantivy temporally.
- set 1 but not 4 threads for inverted text index.
Signed-off-by: chyezh <chyezh@outlook.com>
2024-11-20 20:44:32 +08:00
cqy123456
8216345b07
enhance: reduce copy of bitset and id conversion of brurtforce search ( #37675 )
...
issue: https://github.com/milvus-io/milvus/issues/37798
Signed-off-by: cqy123456 <qianya.cheng@zilliz.com>
2024-11-19 15:48:40 +08:00
Bingyi Sun
6b82320953
fix: Fix using wrong upperbound when searching by pk ( #37769 )
...
issue: https://github.com/milvus-io/milvus/issues/37649
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-19 10:16:31 +08:00
smellthemoon
3d28d99411
fix: to use the correct offset in span ( #37780 )
...
#37734
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2024-11-18 21:56:30 +08:00
aoiasd
16e206167c
enhance: analyzer length filter max should be close interval instead open interval ( #37770 )
...
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-11-18 19:30:31 +08:00
aoiasd
e9391acf80
fix: bm25 brute force search need index params k1 and b ( #37721 )
...
relate: https://github.com/milvus-io/milvus/issues/35853
---------
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-11-18 15:44:31 +08:00
Zhen Ye
3f1614e9d9
enhance: add trace_id into segcore logs ( #37656 )
...
issue: #37655
Signed-off-by: chyezh <chyezh@outlook.com>
2024-11-18 10:20:30 +08:00
aoiasd
3b5a0df159
enhance: Optimize chinese analyzer and support CnAlphaNumFilter ( #37727 )
...
relate: https://github.com/milvus-io/milvus/issues/35853
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-11-16 10:28:30 +08:00
foxspy
0ba868ae64
enhance: update knowhere version ( #37730 )
...
release note draft :
https://github.com/zilliztech/knowhere/releases/tag/v2.5.0
Signed-off-by: xianliang.li <xianliang.li@zilliz.com>
2024-11-16 10:08:30 +08:00
smellthemoon
7999367c0c
fix: use not retried err when get wrong parameter ( #37707 )
...
#37508
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2024-11-15 19:14:30 +08:00
zhagnlu
e4b6773d0a
fix: fix create text index dir conflict bug ( #37693 )
...
#37623
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2024-11-15 18:26:30 +08:00
Bingyi Sun
65d3c6622a
enhance: Optimize GetChunkIDByOffset and add ut ( #37704 )
...
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-15 14:16:31 +08:00
Bingyi Sun
d1596297d9
fix: Fix query failure with inverted index ( #37686 )
...
https://github.com/milvus-io/milvus/issues/37649
---------
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-15 10:28:31 +08:00
foxspy
5ae347aba0
enhance: update knowhere version ( #37688 )
...
issue: #37665 #37631 #37620 #37587 #36906
knowhere has add default nlist value, so some invalid param test ut with
no nlist param will be valid.
Signed-off-by: xianliang.li <xianliang.li@zilliz.com>
2024-11-15 10:10:31 +08:00
Bingyi Sun
1b4f7e3ac1
enhance: Add more expr ut for chunked segment ( #37600 )
...
related pr: #37570
---------
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-14 18:40:32 +08:00
aoiasd
993051bb49
fix: brute force bm25 search lack avgdl param ( #37650 )
...
relate: https://github.com/milvus-io/milvus/issues/35853
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-11-14 14:58:31 +08:00
Buqian Zheng
0565300b7f
fix: Sparse to use CC index as growing/temp index ( #37591 )
...
relate: https://github.com/milvus-io/milvus/issues/35853
Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
2024-11-14 10:54:31 +08:00
aoiasd
1c5b5e1e3d
feat: Add chinese and english analyzer with refactor jieba tokenizer ( #37494 )
...
relate: https://github.com/milvus-io/milvus/issues/35853
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-11-14 10:34:31 +08:00
foxspy
cf883b114e
enhance: update knowhere version ( #37510 )
...
issue: #36925
Signed-off-by: xianliang.li <xianliang.li@zilliz.com>
2024-11-13 16:36:27 +08:00
smellthemoon
3389a6b500
enhance: support null in text match index ( #37517 )
...
#37508
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2024-11-13 11:08:29 +08:00
Zhen Ye
3c225e5c94
fix: data race when using fields_ ( #37612 )
...
issue: #37609
Signed-off-by: chyezh <chyezh@outlook.com>
2024-11-13 04:06:30 +08:00
Chun Han
2d29dcd30c
enhance:refine group_strict_size parameter( #37482 ) ( #37483 )
...
related: #37482
Signed-off-by: MrPresent-Han <chun.han@gmail.com>
Co-authored-by: MrPresent-Han <chun.han@gmail.com>
2024-11-12 09:56:28 +08:00
Bingyi Sun
c1eccce2fa
enhance: enable multiple chunked segment by default ( #37570 )
...
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-12 09:20:28 +08:00
aoiasd
12951f0abb
enhance: rename tokenizer to analyzer and check analyzer params ( #37478 )
...
relate: https://github.com/milvus-io/milvus/issues/35853
---------
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-11-10 16:12:26 +08:00
Bingyi Sun
40ba5a3414
fix: fix chunked segment term filter expression and add ut ( #37392 )
...
issue: https://github.com/milvus-io/milvus/issues/37143
---------
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-07 11:04:19 -08:00
congqixia
5310d3469f
fix: Escape brace of dumped JSON for index err message ( #37504 )
...
Related to #37503
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-11-07 18:58:25 +08:00
smellthemoon
9b6dd23f8e
fix: wrong path spelling when use rootpath in segcore ( #37453 )
...
#36532
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2024-11-07 11:22:25 +08:00
aoiasd
d67853fa89
feat: Tokenizer support build with params and clone for concurrency ( #37048 )
...
relate: https://github.com/milvus-io/milvus/issues/35853
https://github.com/milvus-io/milvus/issues/36751
---------
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-11-06 17:48:24 +08:00
cai.zhang
625b6176cd
fix: Search for pk using raw data to reduce the overhead caused by views ( #37202 )
...
issue: #37152
Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2024-11-05 20:36:24 +08:00
Bingyi Sun
bd04cac4b3
fix: fix group by on chunked segment ( #37292 )
...
https://github.com/milvus-io/milvus/issues/37244
---------
Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-11-05 17:12:23 +08:00
Zhen Ye
9a0e1c82bc
fix: repeated error code in milvus and segcore ( #37359 )
...
issue: #37357
Signed-off-by: chyezh <chyezh@outlook.com>
2024-11-05 16:28:23 +08:00
Zhen Ye
0c4321cf57
fix: crash when startup if the milvus volume is on-operation concurrently ( #37312 )
...
issue: #37311
Signed-off-by: chyezh <chyezh@outlook.com>
2024-11-04 14:50:23 +08:00
foxspy
c27f477b6c
enhance: Update Knowhere version ( #37333 )
...
issue: #37269
Signed-off-by: xianliang.li <xianliang.li@zilliz.com>
2024-11-04 11:56:31 +08:00