Commit Graph

19936 Commits (54797b42860f5917458996b8b5fa734b1f9b15bf)

Author SHA1 Message Date
congqixia 54797b4286
enhance: Refine frequent log in datacoord (#33449)
This PR changes:
- Frequent `ListIndexes` success log to debug level
- Aggregate collection missing log after collection dropped in
`meta.GetCollectionIndexFilesSize`

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-05-29 19:15:43 +08:00
smellthemoon 08b94ea81d
enhance:change wrong log (#33447)
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
2024-05-29 18:35:44 +08:00
congqixia a26d6cdf23
fix: Remove group checker when closing qn pipeline (#33443)
See also #33442

This fix shall prevent group checker keep printing "some node(s) haven't
received input" err message after collection released

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-05-29 14:07:44 +08:00
yihao.dai bbb69980ac
enhance: Replace 'off' with 'disable' (#33433)
YAML will automatically parse "off" as a boolean variable. We should
avoid using "off" in the future.

issue: https://github.com/milvus-io/milvus/issues/32772

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-05-29 12:17:43 +08:00
wei liu b13932bb55
enhance: Enable database level replica num and resource groups for loading collection (#33052)
issue: #30040

This PR introduce two database level props:
1. database.replica.number
2. database.resource_groups

User can set those two database props by AlterDatabase API, then can
load collection without specified replica_num and resource groups. then
it will use database level load param when try to load collections.

Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2024-05-29 10:59:43 +08:00
congqixia 73c9b80a7d
enhance: Store locations for largest K in `LocationCache` (#33429)
See also #32642

`LocationCache` used map to store different locations for different K
which may cause lots of CPU time when get locations many times.

This PR change the implementation of LocationCache to store only the
location for the largest K used to totally remove the map access
operation.

See pprof from test of @XuanYang-cn 

![image](https://github.com/milvus-io/milvus/assets/84113973/ad17cff8-62ad-4d78-9bb0-f6df0512f4ea)

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-05-29 10:05:42 +08:00
sre-ci-robot 371c2d8826 Update all contributors
Signed-off-by: sre-ci-robot <sre-ci-robot@zilliz.com>
2024-05-28 12:00:44 +00:00
jaime 3d29907b6e
enhance: decrease cpu overhead during filter segments on datacoord (#33130)
issue: #33129

Signed-off-by: jaime <yun.zhang@zilliz.com>
2024-05-28 19:17:43 +08:00
Bingyi Sun 6b3e42f8d8
fix: fix wrong default local storage path (#33389)
issue: https://github.com/milvus-io/milvus/issues/33427

Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-05-28 16:49:43 +08:00
Francis b138ae742d
fix: docs for python3 and conan (#32656)
Conan 1.x is broken for Python 3.12, since imp is deprecated in favour
of importlib, removed in 3.12.

https://github.com/milvus-io/milvus/issues/32655

Signed-off-by: light-city <455954986@qq.com>
2024-05-28 14:35:42 +08:00
congqixia e71b7c7cc9
enhance: Reduce datanode metacache frequent scan range (#33400)
See also #32165

There were some frequent scan in metacache:
- List all segments whose start positions not synced
- List compacted segments

Those scan shall cause lots of CPU time when flushed segment number is
large meanwhile `Flushed` segments can be skipped in those two scenarios

This PR make:
- Add segment state shortcut in metacache
- List start positions state before `Flushed`
- Make compacted segments state to be `Dropped` and use `Dropped` state
while scanning them

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-05-28 14:19:42 +08:00
XuanYang-cn 5e39aa9272
enhance: Make channel meta able to writer 200k plus segments (#33279)
See also: #33125

---------

Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2024-05-28 12:33:42 +08:00
SimFG 2453181218
fix: not found database name in the datacoord meta object (#33411)
- issue: #33410

Signed-off-by: SimFG <bang.fu@zilliz.com>
2024-05-28 10:09:48 +08:00
sre-ci-robot 036b68e710 Update all contributors
Signed-off-by: sre-ci-robot <sre-ci-robot@zilliz.com>
2024-05-28 00:01:18 +00:00
ArenaSu a59ed65c48
doc: [skip-e2e] add comments for task queue (#33388)
Add comments for task queue in internal/proxy/task_scheduler.go.

Signed-off-by: ArenaSu <704427617@qq.com>
2024-05-27 20:51:42 +08:00
jaime 0d3272ed6d
enhance: refine logs of cgo pool (#33373)
Signed-off-by: jaime <yun.zhang@zilliz.com>
2024-05-27 19:06:11 +08:00
SimFG 1973ebbc42
fix: quotaCenter collect metrics failed because the wrong param (#33398)
- issue: #33397

Signed-off-by: SimFG <bang.fu@zilliz.com>
2024-05-27 18:17:42 +08:00
aoiasd 59a7a46904
enhance: Merge query stream result for reduce delete task (#32855)
relate: https://github.com/milvus-io/milvus/issues/32854

---------

Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-05-27 18:15:43 +08:00
Ted Xu 066c8ea175
feat: stream reader/writer to support nulls (#33080)
See: #31728

---------

Signed-off-by: Ted Xu <ted.xu@zilliz.com>
2024-05-27 16:27:42 +08:00
wei liu 6275c75013
fix: Watch channel task may stuck forever until qn become offline (#33394)
issue: #32901
pr #32814 introduce the compatible issue, when upgrade to milvus latest,
the query coord may skip update dist due to the lastModifyTs doesn't
changes. but for old version querynode, the lastModifyTs in
GetDataDistritbuionResponse is always 0, which makes qc skip update
dist. then qc will keep retry the task to watch channel again and again.

this PR add compatible with old version querynode, when lastModifyTs is
0, qc will update it's data distribution.

Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2024-05-27 15:01:42 +08:00
nico 5b86226465
test: update test cases about connection (#33257)
Signed-off-by: nico <cheng.yuan@zilliz.com>
2024-05-27 14:59:42 +08:00
Alexander Guzhva 9f9dff33f1
fix: make milvus compilable using gcc-13 (#30149)
add a missing header

Signed-off-by: Alexandr Guzhva <alexanderguzhva@gmail.com>
2024-05-27 14:52:53 +08:00
yihao.dai 5cf4161394
fix: Fix exception info is missing (#33393)
Replace based std::exception to prevent "object slicing"

issue: https://github.com/milvus-io/milvus/issues/33392

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-05-27 14:33:41 +08:00
yihao.dai af71116499
fix: Fix frequent 'failed to get request info' log (#33334)
issue: https://github.com/milvus-io/milvus/issues/33333

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-05-27 14:27:41 +08:00
PowderLi ee73e62202
fix: [restful v2] search result be cut while nq > 1 (#33339)
issue: #33099 #32837 #32419

1. len(search result) may be nq * topk, we need return all rather than
topk
2. the  in restful response payload keep the same with milvus error code

Signed-off-by: PowderLi <min.li@zilliz.com>
2024-05-27 14:25:41 +08:00
SimFG cb99e3db34
enhance: add the includeCurrentMsg param for the Seek method (#33326)
/kind improvement
- issue: #33325

Signed-off-by: SimFG <bang.fu@zilliz.com>
2024-05-27 10:31:41 +08:00
jaime 58ee613fea
enhance: remove repeated stats of loaded entity (#33255)
Signed-off-by: jaime <yun.zhang@zilliz.com>
2024-05-27 01:49:41 +08:00
cai.zhang e708974578
enhance: Remove unused proto importing (#33321)
Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2024-05-27 01:37:28 +08:00
yihao.dai 760223f80a
fix: use seperate warmup pool and disable warmup by default (#33348)
1. use a small warmup pool to reduce the impact of warmup
2. change the warmup pool to nonblocking mode
3. disable warmup by default
4. remove the maximum size limit of 16 for the load pool

issue: https://github.com/milvus-io/milvus/issues/32772

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
Co-authored-by: xiaofanluan <xiaofan.luan@zilliz.com>
2024-05-27 01:25:40 +08:00
Buqian Zheng 1b67cecd65
enhance: add sparse float vector support to restful v2 (#33231)
issue: #29419
also re-enabled an e2e test using restful api, which is previously
disabled due to https://github.com/milvus-io/milvus/issues/32214.

In restful api, the accepted json formats of sparse float vector are:

* `{"indices": [1, 100, 1000], "values": [0.1, 0.2, 0.3]}`
* {"1": 0.1, "100": 0.2, "1000": 0.3}

for accepted indice and value range, see
https://milvus.io/docs/sparse_vector.md#FAQ

Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
2024-05-27 00:47:40 +08:00
Xiaofan 36cbce4def
enhance: optimize datanode cpu usage under large collection number (#33267)
fix #33266 
try to improve cpu usage by refactoring the ttchecker logic and caching
string

Signed-off-by: xiaofanluan <xiaofan.luan@zilliz.com>
2024-05-25 04:43:41 +08:00
zhuwenxing ed883b39d7
test: use ml-dtypes lib to produce bf16 datatype (#33354)
Signed-off-by: zhuwenxing <wenxing.zhu@zilliz.com>
2024-05-24 18:47:41 +08:00
congqixia 970bf18a49
fix: Allocate new slice for each batch in streaming reader (#33359)
Related to #33268

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-05-24 18:07:41 +08:00
Bingyi Sun 370562b4ec
fix: fix partition loaded num metric (#33316)
issue: https://github.com/milvus-io/milvus/issues/32108

Signed-off-by: sunby <sunbingyi1992@gmail.com>
2024-05-24 15:31:42 +08:00
shaoting-huang 3c7d0209d4
enhance: update checker for go version (#33351)
Signed-off-by: shaoting-huang [shaoting-huang@zilliz.com]

issue: https://github.com/milvus-io/milvus/issues/32982

Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
2024-05-24 14:23:40 +08:00
SimFG 2964f60edc
enhance: the proxy metric in the query request (#33307)
/kind improvement
issue: #33306

Signed-off-by: SimFG <bang.fu@zilliz.com>
2024-05-24 14:19:41 +08:00
yiwangdr e895cfed84
fix: reduce redundant map operations in datacoord (#33343)
More refactories will be added.
issue: #33342

Signed-off-by: yiwangdr <yiwangdr@gmail.com>
2024-05-24 12:47:40 +08:00
congqixia 5cdc6ae489
enhance: Sync `deleteBufBytes` config value to default config (#33320)
The delete buffer size is set to 64MB in milvus.yaml but the default set
up shall be 16MB

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-05-24 10:41:40 +08:00
Ted Xu a8bd9bea39
fix: adding blob memory size in binlog serde (#33324)
See: #33280

Signed-off-by: Ted Xu <ted.xu@zilliz.com>
2024-05-24 10:33:40 +08:00
Cai Yudong 4004e4c545
enhance: Optimize bulk insert unittest (#33224)
Issue: #22837

Signed-off-by: Cai Yudong <yudong.cai@zilliz.com>
2024-05-24 10:23:41 +08:00
yihao.dai 7730b910b9
enhance: Decouple compaction from shard (#33138)
Decouple compaction from shard, remove dependencies on shards (e.g.
SyncSegments, injection).

issue: https://github.com/milvus-io/milvus/issues/32809

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-05-24 09:07:41 +08:00
yihao.dai 592d701617
fix: Fix global rate limit is not working (#33335)
If the request is limited by rate limiter, limiter should not "Cancel".
This is because, if limited, tokens are not deducted; instead, "Cancel"
operation would increase the token count.

issue: https://github.com/milvus-io/milvus/issues/31705

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-05-24 09:05:47 +08:00
sre-ci-robot b391781a2e
[automated] Update Knowhere Commit (#33340)
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>
2024-05-24 01:43:40 +08:00
aoiasd 1b4e28b97f
enhance: Check by proxy rate limiter when delete get data by query. (#30891)
relate: https://github.com/milvus-io/milvus/issues/30927

---------

Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-05-23 20:03:40 +08:00
wei liu c7be2ce33a
enhance: Decrease bloom filter fp rate to reduce delete impact (#33301)
when milvus process delete record, it need to find record's corresponded
segment by bloom filter, and higher bloom filter fp rate will cause
delete record forwards to wrong segments.

This PR Decrease bloom filter's default fp to 0.001.

Signed-off-by: Wei Liu <wei.liu@zilliz.com>
2024-05-23 18:15:41 +08:00
zhuwenxing 6c186112bd
test: add sparse vector datatype for import test (#33166)
Signed-off-by: zhuwenxing <wenxing.zhu@zilliz.com>
2024-05-23 17:57:40 +08:00
foxspy 3bec2c4928
enhance: Update Knowhere version (#33309)
/kind branch-feature

Signed-off-by: xianliang.li <xianliang.li@zilliz.com>
2024-05-23 16:57:41 +08:00
congqixia 5452376e90
fix: Remove task from syncmgr after task done (#33302)
See also #33247
Introduced in PR #32865

Remove task after task done to keep checkpoint sound and safe

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-05-23 14:33:40 +08:00
yanliang567 00cd88e164
test: Add index names tests for mulitple vectors support (#33250)
Related issue: https://github.com/milvus-io/milvus/issues/32653
1. Update index name tests
2. remove some time.sleep

---------

Signed-off-by: yanliang567 <yanliang.qiao@zilliz.com>
2024-05-23 14:29:40 +08:00
yihao.dai 155cb40fc9
fix: Fix printing type of request (#33289)
issue: https://github.com/milvus-io/milvus/issues/31705

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-05-23 14:19:40 +08:00