Commit Graph

426 Commits (038eebba4db629edd2498abf1b51009a4b91bf22)

Author SHA1 Message Date
zhagnlu 9489e14000
Optimize multi logical exprs performance when meet some situations (#26265)
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-08-11 15:31:29 +08:00
zhagnlu 411f9ac823
Upgrade minio-go and add region and virtual host config for segcore chunk manager (#26194)
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-08-11 10:37:36 +08:00
Enwei Jiao ca1349708b
Remove time travel ralted testcase (#26119)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-08-10 18:53:17 +08:00
xige-16 1055c90456
Add default retrieve limit (#24782)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2023-08-10 14:11:15 +08:00
yah01 300fef446b
Enable mmap for vector index (#25877)
Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-08-10 13:59:15 +08:00
Jiquan Long 6578a27fe7
Support empty expression (#26153)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2023-08-10 08:37:15 +08:00
Jiquan Long bafb183a2b
Optimize bitset usage (#26096)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2023-08-03 15:25:09 +08:00
MrPresent-Han 5634ba777d
add new threadpool with various priority to avoid deadlock(#25781) (#26028)
Signed-off-by: MrPresent-Han <chun.han@zilliz.com>
2023-08-03 09:31:07 +08:00
Jiquan Long 5c1f79dc54
Push down the limit operator to segcore (#25959)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2023-08-01 20:29:05 +08:00
yah01 9618bd9b42
Set channel capacity before consuming it (#25895)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-07-26 17:35:01 +08:00
foxspy 5437fcce8e
fix range search (#25880)
Signed-off-by: xianliang <xianliang.li@zilliz.com>
2023-07-25 10:07:01 +08:00
yah01 53c3bf053e
Fix unstable sealed segment bruteforce unittest (#25867)
Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-07-25 09:05:00 +08:00
Cai Yudong 9a4761dcc7
Remove binary metrics TANIMOTO/SUPERSTRUCTURE/SUBSTRUCTURE (#25708)
Signed-off-by: Yudong Cai <yudong.cai@zilliz.com>
2023-07-19 16:16:58 +08:00
zhagnlu 65cb52d06b
Support dynamic simd framework and using term expr as example (#25260)
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-07-13 16:22:30 +08:00
yah01 dd5f896dc8
Load batch by batch (#25212)
This will significantly reduce the memory usage while loading
- 1x memory usage and MBs overhead for buffer (memory mode)
- only MBs overhead for buffer (mmap mode)

Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-07-06 13:58:27 +08:00
Jiquan Long c2a2ece551
Fix expression on integer overflow case (#25320)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2023-07-05 10:18:24 +08:00
foxspy 31173727b2
growing segment index memory opt & get vector bugfix (#25272)
Signed-off-by: xianliang <xianliang.li@zilliz.com>
2023-07-05 00:04:25 +08:00
Enwei Jiao 816158e4af
Remove outdated searchplan (#25282)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-07-04 18:30:25 +08:00
xige-16 04082b3de2
Migrate the ability to upload and download binlog to cpp (#22984)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2023-06-25 14:38:44 +08:00
PowderLi 3f4356df10
fix the spelling of `field` (#25008)
Signed-off-by: PowderLi <min.li@zilliz.com>
2023-06-21 14:00:42 +08:00
yah01 a413842e38
Fix deleted data is still visible (#24849)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-06-16 17:16:41 +08:00
zhagnlu c5b1533fdc
Change segcore search_id from travelling all bits to select true bits(#24659) (#24800)
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-06-16 16:48:44 +08:00
cqy123456 a519213316
Update knowhere version, update diskann api and generate cache nodes in build process (#24898)
Signed-off-by: cqy123456 <qianya.cheng@zilliz.com>
2023-06-16 14:20:39 +08:00
Jiquan Long 26a6e1b946
Fix BinaryRange expression on integer overflow case (#24920)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2023-06-16 09:56:39 +08:00
zhagnlu f60b839127
Support element in json array in segcore part(#24677) (#24829)
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-06-14 14:38:37 +08:00
Jiquan Long c6a6998ba7
Fix term expression on interger overflow case (#24867)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2023-06-14 11:44:38 +08:00
Cai Yudong 1b3c4b26f1
Optimize range search result sort in segcore (#24837)
Signed-off-by: Yudong Cai <yudong.cai@zilliz.com>
2023-06-13 19:22:38 +08:00
yah01 ceda0ed598
Optimize the performance of filter by JSON field (#24268)
- Construct JSON pointer only once
- Avoid copying nested path for each row

Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-05-22 00:47:25 +08:00
yah01 c75e7a5d05
Fix failed to compare int value with double value (#24229)
Signed-off-by: yah01 <yah2er0ne@outlook.com>
2023-05-19 12:57:23 +08:00
zhagnlu 113f9a0ebc
Support SIMD of several Expr (#23715) (#23717)
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-05-12 14:11:20 +08:00
cai.zhang 9715a850fa
Support expr with json field (#23804)
Signed-off-by: cai.zhang <cai.zhang@zilliz.com>
2023-05-10 10:19:19 +08:00
yah01 62eea5286f
Support to filter with json expr (#23739)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-04-30 20:36:39 +08:00
yah01 60fdd7e4f4
Introduce simdjson (#23644)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-04-26 10:30:34 +08:00
foxspy 6f4ed517de
add growing segment index (#23615)
Signed-off-by: xianliang <xianliang.li@zilliz.com>
2023-04-26 10:14:41 +08:00
yihao.dai 092d743917
Add support for getting vectors by ids (#23450)
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2023-04-23 09:00:32 +08:00
yah01 546080dcdd
Support to retrieve json (#23563)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-04-21 11:46:32 +08:00
Cai Yudong 5f4673fd16
Optimize unittest to save runtime (#23248)
Signed-off-by: Yudong Cai <yudong.cai@zilliz.com>
2023-04-07 14:20:29 +08:00
Cai Yudong ef63e64ded
Remove ANNOY index type (#23189)
Signed-off-by: Yudong Cai <yudong.cai@zilliz.com>
2023-04-04 16:30:27 +08:00
Enwei Jiao 50dab2d394
Add tracing for retrieve (#23033)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-03-29 18:10:02 +08:00
Enwei Jiao ef57ba3bac
Remove easylogging and replace by GLog (#22979)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-03-27 18:22:02 +08:00
yah01 081572d31c
Refactor QueryNode (#21625)
Signed-off-by: yah01 <yang.cen@zilliz.com>
Co-authored-by: Congqi Xia <congqi.xia@zilliz.com>
Co-authored-by: aoiasd <zhicheng.yue@zilliz.com>
2023-03-27 00:42:00 +08:00
Cai Yudong 0e9a4478e3
Remove useless index mode (#22934)
Signed-off-by: Yudong Cai <yudong.cai@zilliz.com>
2023-03-23 21:39:59 +08:00
Enwei Jiao fd974df09f
OTEL integration in cpp (#22175)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-03-21 16:31:57 +08:00
zhagnlu 7c633e9b9d
Support aliyun OSS for chunk manager (#22663) (#22842)
Signed-off-by: luzhang <luzhang@zilliz.com>
Co-authored-by: luzhang <luzhang@zilliz.com>
2023-03-20 14:45:57 +08:00
xige-16 9aa99aedbb
[Cherry-Pick] Remove arrow uasge in FieldData (#22726)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2023-03-20 10:41:56 +08:00
Jiquan Long 8139106b51
Feat: count entities by expression (#22765)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2023-03-16 19:31:55 +08:00
Cai Yudong ab3cbdfc61
Partial change to prepare for GPU index type support (#22591)
Signed-off-by: Yudong Cai <yudong.cai@zilliz.com>
2023-03-14 23:21:56 +08:00
Jiquan Long a36fefb009
Fix cpplint (#22657)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2023-03-10 09:47:54 +08:00
XuanYang-cn 955bc06165
Fix search binary pagination failure (#22477)
See also: #22168

Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2023-03-08 11:03:51 +08:00
smellthemoon 0dae20c938
Layer range search results (#22539)
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2023-03-03 15:19:48 +08:00
yah01 7478e44911
Support using mmap to load data (#22052)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-03-01 18:07:49 +08:00
smellthemoon 8d05874ff9
Fix get index mode error (#22364)
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2023-02-27 14:45:47 +08:00
smellthemoon 820d0bc060
Fix range search error (#22352)
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2023-02-23 14:21:46 +08:00
smellthemoon 9e0ec15436
Support range search (#21652)
Signed-off-by: smellthemoon <xinguo.li@zilliz.com>
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: jaime <yun.zhang@zilliz.com>
2023-02-21 09:48:32 +08:00
xige-16 8ec5ba13dc
Fix search failed on disk index when search_list equal to limit (#21113)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2023-02-16 11:02:34 +08:00
presburger 9950cacd10
support knowhere 2.0 (#21857)
Signed-off-by: Yusheng.Ma <Yusheng.Ma@zilliz.com>
2023-02-10 14:24:32 +08:00
yah01 73ce87dfe5
Fix reduce decreasing recall (#21981)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-02-06 11:23:53 +08:00
yah01 836773f1a0
Optimize performance of reducing segments (#21722)
- Improve the performance of reducing from O(knlogn) to O(nlogk)

Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-01-16 15:25:42 +08:00
smellthemoon bf3c02155a
Support update (#20875)
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2023-01-04 17:21:36 +08:00
Jiquan Long d7156812c1
Try using ASAN in ci ut (#21089)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2022-12-29 15:29:31 +08:00
smellthemoon 1e5cd9dc6e
Modify comment (#21362)
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2022-12-26 15:27:30 +08:00
xige-16 a20770c172
Delete logs that print sensitive information (#20889)
Signed-off-by: xige-16 <xi.ge@zilliz.com>

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-12-01 10:35:16 +08:00
Enwei Jiao 958e94f6f0
Use Conan as c++ package manager (#19920)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>

Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2022-11-23 10:39:11 +08:00
xige-16 34c88cea32
Fix GetDiskUsedSize not work (#20474)
Signed-off-by: xige-16 <xi.ge@zilliz.com>

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-11-21 17:47:17 +08:00
smellthemoon f5ab719f21
timestamp decided if the pks were the same (#20166)
Signed-off-by: lixinguo <xinguo.li@zilliz.com>

Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2022-11-21 10:55:10 +08:00
xige-16 4a66965df4
Delete RAW_DATA copy when load IVF_FLAT index data (#20274)
Signed-off-by: xige-16 <xi.ge@zilliz.com>

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-11-05 17:33:05 +08:00
zhagnlu cb2591d1fe
Parallel operations of disk-ann file manager (#20120) (#20122)
Signed-off-by: zhagnlu <lu.zhang@zilliz.com>

Signed-off-by: zhagnlu <lu.zhang@zilliz.com>
Co-authored-by: zhagnlu <lu.zhang@zilliz.com>
2022-11-03 14:39:40 +08:00
zhagnlu 02a8a07322
Fix BucketExists bug in minio chunk manager (#19548) (#19854)
Signed-off-by: zhagnlu <lu.zhang@zilliz.com>

Signed-off-by: zhagnlu <lu.zhang@zilliz.com>
Co-authored-by: zhagnlu <lu.zhang@zilliz.com>
2022-10-20 10:23:27 +08:00
xige-16 158787811e
Move assemble/disassemble func to core (#19420)
Signed-off-by: xige-16 <xi.ge@zilliz.com>

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-10-16 21:17:25 +08:00
xige-16 a1db9038fb
Move disk index params to config file (#19714)
Signed-off-by: xige-16 <xi.ge@zilliz.com>

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-10-14 17:51:24 +08:00
xige-16 8c9c1672ae
Assign different storage config for indexes (#19517)
Signed-off-by: xige-16 <xi.ge@zilliz.com>

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-10-14 14:45:23 +08:00
congqixia 089912baa7
Fix segcore unit test name (#19783)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-10-14 14:29:24 +08:00
Cai Yudong 87d78a4a85
Ignore cases when comparing metric type in segcore (#19437)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-09-26 17:58:52 +08:00
aoiasd 2b58bd5c0a
Optimize large memory usage of InsertRecord by using vector instead of unordered_map if InsertRecord used in sealed segment (#19245)
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>

Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2022-09-23 17:08:51 +08:00
xige-16 428840178c
Support diskann index for vector field (#19093)
Signed-off-by: xige-16 <xi.ge@zilliz.com>

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-09-21 20:16:51 +08:00
aoiasd 58d3a49b62
Optimize large memory usage of Variant used in InsertRecord (#19197)
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2022-09-16 10:52:47 +08:00
xige-16 4de1bfe5bc
Add cpp data codec (#18538)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
Co-authored-by: zhagnlu lu.zhang@zilliz.com

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-09-09 22:12:34 +08:00
Cai Yudong 7b5eb85205
Remove redundant parameter for segcore Search() (#19025)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-09-06 12:05:11 +08:00
Cai Yudong 765907ab77
Optimize segcore Reduce (#18902)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-09-06 10:55:12 +08:00
Cai Yudong 686b0ce796
Upgrade to knowhere-v1.3.0, remove following index support: (#18935)
- IVF_SQ8H
- RHNSW_FLAT/RHNSW_PQ/RHNSW_SQ
- NGT
- NSG
- SPTAG

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-09-05 10:41:11 +08:00
Cai Yudong da96659569
[skip e2e] Add more testcases with different parameter combinations in test_reduce (#18967)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-09-01 16:59:02 +08:00
Cai Yudong 4ded45394a
[skip e2e] Optimize test_reduce (#18957)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-09-01 11:09:00 +08:00
Cai Yudong 9dc3bbecbd
Change the type of slice_nqs and slice_topks from int32_t[] to int64_t[] (#18867)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-08-29 11:36:56 +08:00
Letian Jiang c6d9d0df5e
Fix segcore benchmark (#18675)
Signed-off-by: Letian Jiang <letian.jiang@zilliz.com>

Signed-off-by: Letian Jiang <letian.jiang@zilliz.com>
2022-08-17 20:24:49 +08:00
Cai Yudong d878cc2ec0
Rename real_topK_per_nq_ to topks_per_nq_prefix_sum_ in SearchResult struct (#18634)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>

Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-08-16 00:32:48 +08:00
zhenshan.cao a287a2b3fd
Return empty result in advance if all data filtered out (#18329) (#18438)
Signed-off-by: zhenshan.cao <zhenshan.cao@zilliz.com>
2022-07-28 12:36:30 +08:00
bigsheeper cef8b1e7cc
Enable jemalloc (#18349)
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2022-07-20 22:22:31 +08:00
Jeng.Gwan 638f6c36e9
Support to get real row count of segment (#18115)
Signed-off-by: xaxys <zheng.guan@zilliz.com>
2022-07-18 09:58:28 +08:00
Cai Yudong 015a2f0866
Let growing segment call knowhere brute search API (#18227)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-07-12 11:58:25 +08:00
bigsheeper 22508f36d3
Get search field id from search plan and log if loaded index when search segments (#18183)
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2022-07-08 20:18:22 +08:00
zhagnlu 257da153ce
Fix core dump when nq has no topk result (#17923) (#18051)
Signed-off-by: zhagnlu <lu.zhang@zilliz.com>

Co-authored-by: zhagnlu <lu.zhang@zilliz.com>
2022-07-05 19:48:20 +08:00
xige-16 0702993106
Fix query not empty after delete string (#17953)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-07-04 16:54:20 +08:00
Jiquan Long 6954a5ba3e
Fix search successfully with invalid metric type (#17977)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2022-07-01 22:28:23 +08:00
bigsheeper 641fb96958
Return malloc_trim result and log if malloc_trim successfully (#17959)
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2022-07-01 10:44:18 +08:00
Cai Yudong a001412e12
Replace faiss::MetricType with knowhere::MetricType (#17891)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-06-29 14:20:19 +08:00
Letian Jiang 4ae1ca2cac
Fix arrow builder nullptr check in FinishPayloadWriter (#17873)
Signed-off-by: Letian Jiang <letian.jiang@zilliz.com>
2022-06-28 20:04:17 +08:00
Enwei Jiao 16c3aedc15
refine complie configuration (#17502)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2022-06-24 21:12:15 +08:00
Cai Yudong 7385770014
Upgrade to knowhere-v1.1.12 (#17692)
Signed-off-by: yudong.cai <yudong.cai@zilliz.com>
2022-06-24 10:34:18 +08:00
bigsheeper 1722431699
Checking for existence in map in segcore (#17700)
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2022-06-22 18:16:14 +08:00
Jiquan Long fdccfb4273
Fix ut heap overflow (#17706)
Signed-off-by: longjiquan <jiquan.long@zilliz.com>
2022-06-22 16:56:13 +08:00