Commit Graph

412 Commits (cdc_test)

Author SHA1 Message Date
PowderLi 4feb3fa7c6
support azure (#26398)
Signed-off-by: PowderLi <min.li@zilliz.com>
2023-09-19 10:01:23 +08:00
Xu Tong 9166011c4a
Add float16 vector (#25852)
Signed-off-by: Writer-X <1256866856@qq.com>
2023-09-08 10:03:16 +08:00
bjzhjing 548c82eca5
Refactor storage.MergeInsertData() to optimize the merging process (#26839)
Benchmark Milvus with https://github.com/qdrant/vector-db-benchmark and
specify the datasets as 'deep-image-96-angular'. Meanwhile, do perf
profiling during 'upload + index' stage of vector-db-benchmark and see
the following hot spots.

39.59%--github.com/milvus-io/milvus/internal/storage.MergeInsertData
        |
        |--21.43%--github.com/milvus-io/milvus/internal/storage.MergeFieldData
        |          |
        |          |--17.22%--runtime.memmove
        |                     |
        |                     |--1.53%--asm_exc_page_fault
        |                     ......
        |
        |--18.16%--runtime.memmove
                   |
                   |--1.66%--asm_exc_page_fault
                   ......

The hot code path is in storage.MergeInsertData() which updates
buffer.buffer by creating a new 'InsertData' instance and merging both
the old buffer.buffer and addedBuffer into it. When it calls golang
runtime.memmove to move buffer.buffer which is with big size (>1M), the
hot spots appear.

To avoid the above overhead, update storage.MergeInsertData() by
appending addedBuffer to buffer.buffer, instead of moving buffer.buffer
and addedBuffer to a new 'InsertData'. This change removes the hot spots
'runtime.memmove' from perf profiling output. Additionally, the 'upload
+ index' time, which is one performance metric of vector-db-benchmark,
is reduced around 60% with this change.

Signed-off-by: Cathy Zhang <cathy.zhang@intel.com>
2023-09-05 21:41:48 +08:00
Enwei Jiao fb0705df1b
Decouple basetable and componentparam (#26725)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-09-05 10:31:48 +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
congqixia 2770ac4df5
Fix nilness linter errors (#26218)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-08-09 11:31:15 +08:00
zhenshan.cao 2c6c7749e2
Enable print_log support json data type (#26118)
Signed-off-by: zhenshan.cao <zhenshan.cao@zilliz.com>
2023-08-04 11:27:05 +08:00
xige-16 f33451b3d8
Write the cache file to the cacheStorage.rootpath dir (#25715)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2023-07-28 10:59:02 +08:00
xige-16 94d6cbb238
Fix querynode panic when binlog ts wrong (#25635)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2023-07-18 10:41:20 +08:00
xige-16 33c2012675
Add more metrics (#25081)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2023-06-26 17:52:44 +08:00
Xiaofan e8911ebda7
Add retry time when lazy load BF (#25096)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2023-06-25 11:32:43 +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 8bc5282eb3
Fix datanode always retries to load stats even file corrupted (#25012)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-06-20 16:40:42 +08:00
Enwei Jiao 1ef8f0fceb
Remove cgo PayloadWriter (#24892)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-06-14 18:04:38 +08:00
yah01 a9dccec03a
Add go payload writer (#24656) (#24762)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-06-09 13:52:39 +08:00
congqixia 41af0a98fa
Use go-api/v2 for milvus-proto (#24770)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-06-09 01:28:37 +08:00
yah01 ebd0279d3f
Check error by Error() and NoError() for better report message (#24736)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-06-08 15:36:36 +08:00
Enwei Jiao d3af451d92
Upgrade golangci-lint (#24707)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-06-07 19:34:36 +08:00
Bingyi Sun b71c967ed7
Fix NoSuchKey error caused by special stats log (#24670)
Signed-off-by: sunby <bingyi.sun@zilliz.com>
Co-authored-by: sunby <bingyi.sun@zilliz.com>
2023-06-06 17:34:36 +08:00
aoiasd c84bdcea49
merge stats log when segment flushing or compacting (#23570)
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2023-05-29 10:21:28 +08:00
congqixia 73a181d226
Fix get vector it timeout and improve some string const usage (#24141)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-05-16 17:41:22 +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
Enwei Jiao 967a97b9bd
Support json & array types (#23408)
Signed-off-by: yah01 <yang.cen@zilliz.com>
Co-authored-by: yah01 <yang.cen@zilliz.com>
2023-04-20 11:32:31 +08:00
cai.zhang 9288020da3
Fix bug for batch delete files on gcp of minio (#23052) (#23083) (#23090)
Signed-off-by: cai.zhang <cai.zhang@zilliz.com>
2023-04-09 16:22:30 +08:00
jaime c9d0c157ec
Move some modules from internal to public package (#22572)
Signed-off-by: jaime <yun.zhang@zilliz.com>
2023-04-06 19:14:32 +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
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
huanghaoyuanhhy 024beddfe6
Make GCS OAuth token thread-safe (#22714)
Signed-off-by: huanghaoyuan <haoyuan.huang@zilliz.com>
2023-03-14 18:09:54 +08:00
congqixia 732986aa04
Remove fmt.Print from internal package (#22722)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-03-14 17:36:05 +08:00
shaoyue 32581e6452
Support aliyun oss as object storage with ak or IAM (#22376)
Signed-off-by: shaoyue.chen <shaoyue.chen@zilliz.com>
2023-03-09 16:33:52 +08:00
yah01 90a5aa6265
Refine errors, re-define error codes (#22501)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2023-03-09 15:47:52 +08:00
jaime d126f06946
Decouple mq module from internal proto definition (#22536)
Signed-off-by: jaime <yun.zhang@zilliz.com>
2023-03-04 23:21:50 +08:00
congqixia ea4ab9e4a4
Replace ErrorList with multiErrors (#22532)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-03-03 14:59:49 +08:00
Enwei Jiao 697dedac7e
Use cockroachdb/errors to replace other error pkg (#22390)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2023-02-26 11:31:49 +08:00
jaime 5936723904
Refine Read method of MinioChunkManager (#22235)
Signed-off-by: jaime <yun.zhang@zilliz.com>
2023-02-21 16:22:26 +08:00
congqixia f2575e5fa8
Add unconvert & durationcheck linters and fix issues (#22161)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-02-15 17:22:34 +08:00
Xiaofan 6004867853
Add bucket name in error message (#21887)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2023-02-10 17:44:31 +08:00
Xiaofan 949d5d078f
Fix memory calculation in dataCodec (#21800)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2023-01-28 11:09:52 +08:00
congqixia 5986106037
Make paramtable init only once (#21782)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2023-01-19 14:53:44 +08:00
Enwei Jiao 264f08953d
Refactor Rocksmq's config (#21304)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2022-12-26 19:11:30 +08:00
Enwei Jiao 166e9f0da5
Refactor GrpcConfig (#21142)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>

Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2022-12-16 15:59:23 +08:00
Xiaofan 908023a06e
Add more test on file reader (#21082)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>

Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-12-09 11:55:19 +08:00
cai.zhang eb7ef01b9a
Fix bug for ListWithPrefix from minio with recursive is false (#21059)
Signed-off-by: cai.zhang <cai.zhang@zilliz.com>

Signed-off-by: cai.zhang <cai.zhang@zilliz.com>
2022-12-08 15:47:18 +08:00
Enwei Jiao 89b810a4db
Refactor all params into ParamItem (#20987)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>

Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2022-12-07 18:01:19 +08:00
jaime 548e90ec68
Support an analogous Java Guava cache implementation (#20831)
Signed-off-by: yun.zhang <yun.zhang@zilliz.com>

Signed-off-by: yun.zhang <yun.zhang@zilliz.com>
2022-12-05 20:29:18 +08:00
congqixia f745d7f489
Fix compaction target segment rowNum is always 0 (#20937)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-12-01 20:33:17 +08:00
yah01 f76ea292d2
Change LRU to a generic type (#20744)
Signed-off-by: yah01 <yang.cen@zilliz.com>

Signed-off-by: yah01 <yang.cen@zilliz.com>
2022-11-22 12:11:11 +08:00
groot bf2107ecf5
Fix a regression of local storage (#20653)
Signed-off-by: yhmo <yihua.mo@zilliz.com>

Signed-off-by: yhmo <yihua.mo@zilliz.com>
2022-11-21 10:19:10 +08:00
Xiaofan 633a749880
Recude IndexCodec Load Memory (#20621)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>

Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-11-18 10:47:08 +08:00
Enwei Jiao c05b9ad539
Add event dispatcher for config (#20393)
Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>

Signed-off-by: Enwei Jiao <enwei.jiao@zilliz.com>
2022-11-17 18:59:09 +08:00
shaoyue 09ea38615e
Fix gcp oauth token not cached (#20380)
Signed-off-by: shaoyue.chen <shaoyue.chen@zilliz.com>

Signed-off-by: shaoyue.chen <shaoyue.chen@zilliz.com>
2022-11-08 12:43:02 +08:00
Xiaofan 766957e75a
Recuce memory consumption in MinioChunkManager (#20284)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>

Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-11-03 12:19:35 +08:00
shaoyue 9fac1476f3
Add support for GCS(GoogleCloudStorage) with IAM (#20164)
go mod tidy

Signed-off-by: shaoyue.chen <shaoyue.chen@zilliz.com>

Signed-off-by: shaoyue.chen <shaoyue.chen@zilliz.com>
2022-11-01 11:07:35 +08:00
Xiaofan 2bfecf5b4e
Refine bloomfilter and memory usage (#20168)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>

Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-10-31 17:41:34 +08:00
SimFG a55f739608
Separate public proto files (#19782)
Signed-off-by: SimFG <bang.fu@zilliz.com>

Signed-off-by: SimFG <bang.fu@zilliz.com>
2022-10-16 20:49:27 +08:00
yah01 89c9cb3680
Fix scan may break GC limitation (#19670)
Signed-off-by: yah01 <yang.cen@zilliz.com>

Signed-off-by: yah01 <yang.cen@zilliz.com>
2022-10-15 15:13:24 +08:00
jaime 1919353f02
Fix parse segment id error from delta log (#19566)
Signed-off-by: yun.zhang <yun.zhang@zilliz.com>

Signed-off-by: yun.zhang <yun.zhang@zilliz.com>
2022-09-30 14:18:55 +08:00
congqixia 838a633584
Add ctx parameter for ChunkManager methods (#19546)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-09-29 16:18:56 +08:00
Ten Thousand Leaves b30c9d4f60
Support Bulk Load in Milvus 2.2 (#18982)
Signed-off-by: Yuchen Gao <yuchen.gao@zilliz.com>

Signed-off-by: Yuchen Gao <yuchen.gao@zilliz.com>
2022-09-26 18:06:54 +08:00
Xiaofan 928a213e31
Fix Chunk Manager Path mis use (#19141)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>

Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-09-23 14:40:51 +08:00
congqixia f24e16a66f
Handle multiple parquet RowGroup/Column (#19283)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-09-21 09:12:50 +08:00
SimFG d7f38a803d
Separate some proto files (#19218)
Signed-off-by: SimFG <bang.fu@zilliz.com>

Signed-off-by: SimFG <bang.fu@zilliz.com>
2022-09-16 16:56:49 +08:00
groot b161aec95e
Support input empty string (#19111) (#19144)
Signed-off-by: yhmo <yihua.mo@zilliz.com>

Signed-off-by: yhmo <yihua.mo@zilliz.com>

Signed-off-by: yhmo <yihua.mo@zilliz.com>
2022-09-13 13:36:29 +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
congqixia ce9ba0dd02
Remove debug log in minio chunk manager (#19134)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-09-09 17:32:37 +08:00
congqixia db23a256c2
Make minioChunkManager ListObject level by level when recursive is true (#19096)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-09-08 14:58:34 +08:00
congqixia 68a6587374
Set insert&stats binlog timestamp range (#19005)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-09-04 09:05:09 +08:00
congqixia c91bb0b015
Unify rootPath in configs and ChunkManager (#18808)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-08-25 19:32:53 +08:00
codeman 70254c3ec5
Unified catalog interface for segment (#18289) (#18290)
Signed-off-by: kejiang <ke.jiang@zilliz.com>

Signed-off-by: kejiang <ke.jiang@zilliz.com>
Co-authored-by: kejiang <ke.jiang@zilliz.com>
2022-08-20 10:24:51 +08:00
xige-16 e40061b864
Update binlog event format (#18347)
Signed-off-by: xige-16 <xi.ge@zilliz.com>

Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-08-11 14:06:38 +08:00
jaime efa5dfaa7b
Fix Parse binlog path failure (#18583)
Signed-off-by: yun.zhang <yun.zhang@zilliz.com>
2022-08-10 17:04:37 +08:00
jaime 6e7831470c
Revert settings of Pulsar address and refine config code (#18494)
Signed-off-by: yun.zhang <yun.zhang@zilliz.com>
2022-08-02 21:26:33 +08:00
jaime c9174d55ba
Refine merge operation during compacting phase (#18399)
Signed-off-by: yun.zhang <yun.zhang@zilliz.com>
2022-07-28 14:52:31 +08:00
congqixia e9d777b336
Fix ParseSegmentIDBinlog panicks with bad input (#18413)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-07-26 19:32:30 +08:00
xing.zhao d90313a1a8
Use chunkManager rather than minio.Client in datanode garbage collection (#18092)
Signed-off-by: xingzhao <xing.zhao@zilliz.com>

Co-authored-by: xingzhao <xing.zhao@zilliz.com>
2022-07-22 22:10:28 +08:00
Xiaofan 6d82ef8c20
fix garbage collector err handling (#18277)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-07-15 13:54:26 +08:00
cai.zhang 0ad15a22c8
IndexCoord handle events correctly (#17878)
Signed-off-by: Cai.Zhang <cai.zhang@zilliz.com>
2022-07-07 14:44:21 +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
congqixia 37464a281d
Fix datasorter swap vector logic (#17661)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-06-21 11:38:12 +08:00
shaoyue 04f836f516
Fix s3 delete file (#17574)
Signed-off-by: shaoyue.chen <shaoyue.chen@zilliz.com>
2022-06-16 18:14:10 +08:00
yah01 70f8bea4b4
Avoid growing slice as deserializing binlogs (#17421)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2022-06-08 11:46:06 +08:00
shaoyue 76eaa3fc50
Support access s3 through AWS IAM role (#17292)
Signed-off-by: shaoyue.chen <shaoyue.chen@zilliz.com>
2022-06-02 19:42:03 +08:00
Enwei Jiao d28a2db46c
move arrow from storage to core (#17061)
Signed-off-by: Enwei Jiao <jiaoew2011@gmail.com>
2022-05-22 20:03:58 +08:00
Xiaofan 5355153805
Minio Error is not Handled gracefully (#17003)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-05-16 19:25:55 +08:00
Letian Jiang 72bbe40254
Make PayloadReader read column data in batch (#16826)
Signed-off-by: Letian Jiang <letian.jiang@zilliz.com>
2022-05-10 11:37:52 +08:00
cai.zhang 2be46a01c0
Reduce the number of retries and add error log (#16754)
Signed-off-by: cai.zhang <cai.zhang@zilliz.com>
2022-05-05 09:31:51 +08:00
Letian Jiang 5b2b917987
Benchmark go/cgo parquet payload readers (#16736)
Signed-off-by: Letian Jiang <letian.jiang@zilliz.com>
2022-04-29 15:59:47 +08:00
xige-16 515d0369de
Support string type in segcore (#16546)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
Co-authored-by: dragondriver <jiquan.long@zilliz.com>

Co-authored-by: dragondriver <jiquan.long@zilliz.com>
2022-04-29 13:35:49 +08:00
jaime 68b1b82faf
Remove DataKV interface (#16692)
Signed-off-by: yun.zhang <yun.zhang@zilliz.com>
2022-04-28 21:03:47 +08:00
yah01 7af02fa531
Improve load performance, load binlogs concurrently per file, deserialize binlogs concurrently per field/segment (#16514)
Signed-off-by: yah01 <yang.cen@zilliz.com>
2022-04-25 15:57:47 +08:00
Ten Thousand Leaves 4ef2df8cb9
Fix an issue where query coord calls MinioChunkManager during init (#16629)
It causes Milvus not runnable with local storage.

issue: #15604

/kind enhancement

Signed-off-by: Yuchen Gao <yuchen.gao@zilliz.com>
2022-04-25 11:11:46 +08:00
congqixia 9769426e84
Use fmt.Errorf instead of string concat in local_chunk_manager.go (#16442)
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2022-04-19 18:39:39 +08:00
Ten Thousand Leaves e76a8c5ec2
Update compile options and config for embedded Milvus (#16472)
/kind enhancement

issue: #15711
Signed-off-by: Yuchen Gao <yuchen.gao@zilliz.com>
2022-04-14 19:57:34 +08:00
godchen 4781db8a2a
Add datanode import (#16414)
Signed-off-by: godchen0212 <qingxiang.chen@zilliz.com>
2022-04-12 22:19:34 +08:00
godchen bb7a0766fe
Add dependency factory (#16204)
Signed-off-by: godchen0212 <qingxiang.chen@zilliz.com>
2022-04-07 22:05:32 +08:00
xige-16 99984b88e1
Support delete varChar value (#16229)
Signed-off-by: xige-16 <xi.ge@zilliz.com>
2022-04-02 17:43:29 +08:00
Jiquan Long ba37531456
Add support for loading multiple indexes (#16138)
Signed-off-by: dragondriver <jiquan.long@zilliz.com>
2022-03-30 21:11:28 +08:00
Xiaofan 801eeffbcc
Replace cgo parquet reader to go parquet reader (#16199)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-03-30 15:21:28 +08:00
Xiaofan b6b3c9863c
Fix incompatibility of stats log (#16277)
Signed-off-by: xiaofan-luan <xiaofan.luan@zilliz.com>
2022-03-30 10:15:28 +08:00