In quota center, ignore the "DB not found error" to prevent it from
affecting the rate limiting of other databases.
/kind improvement
---------
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
1. support isClusteringKey in restful api;
2. throw err if passed invalid 'enableDynamicField' params
3. parameters in indexparams are not processed properly, related with
#36365
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
issue: https://github.com/milvus-io/milvus/issues/35853
* BM25 Function now takes no params, k1, b should be passed via index
params
* support BM25 full text search when metric type is not present in
search request
* add more strict validation with functions at collection creation time
Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
This PR splits sealed segment to chunked data to avoid unnecessary
memory copy and save memory usage when loading segments so that loading
can be accelerated.
To support rollback to previous version, we add an option
`multipleChunkedEnable` which is false by default.
Signed-off-by: sunby <sunbingyi1992@gmail.com>
Support the new RESTful URL for retrieving/describing import progress:
`/v2/vectordb/jobs/import/describe`.
Deprecate the old URL: `/v2/vectordb/jobs/import/get_progress`.
issue: https://github.com/milvus-io/milvus/issues/36752
---------
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
issue: #35922
add an enable_tokenizer param to varchar field: must be set to true so
that a varchar field can enable_match or used as input of BM25 function
---------
Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
See #36550
This PR made 2 changes:
1. Introducing a prioritization mechanism, if
`dataCoord.compaction.taskPrioritizer` is set to `level`, compaction
tasks are always executed as the priority of L0>Mix>Clustering
2. `dataCoord.compaction.maxParallelTaskNum` now controls the
parallelism of executing tasks, not the task number of queue +
executing.
---------
Signed-off-by: Ted Xu <ted.xu@zilliz.com>
1. Optimize import scheduling strategic:
a. Revise slot weights, calculating them based on the number of files
and segments for both import and pre-import tasks.
b. Ensure that the DN executes tasks in ascending order of task ID.
2. Add time cost metric and log.
issue: https://github.com/milvus-io/milvus/issues/36600,
https://github.com/milvus-io/milvus/issues/36518
---------
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
Directly add import segments from the meta, eliminating the dependency
on the segment manager.
issue: https://github.com/milvus-io/milvus/issues/34648
---------
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
issue: #36464
This PR enable balance on querynode with different mem capacity, for
query node which has more mem capactity will be assigned more records,
and query node with the largest difference between assignedScore and
currentScore will have a higher priority to carry the new segment.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Previous label case broken by #36107, this PR make all inbound label
using label constants from metrics package.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Native support for Google cloud storage using the Google Cloud Storage
libraries. Authentication is performed using GCS service account
credentials JSON.
Currently, Milvus supports Google Cloud Storage using S3-compatible APIs
via the AWS SDK. This approach has the following limitations:
1. Overhead: Translating requests between S3-compatible APIs and GCS can
introduce additional overhead.
2. Compatibility Limitations: Some features of the original S3 API may
not fully translate or work as expected with GCS.
To address these limitations, This enhancement is needed.
Related Issue: #36212
issue: #36490
After the query node changes from a delegator to a worker, proxy should
skip this querynode's health check.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: #36488
when call TransferChannel/TransferSegment, querycoord will generate and
submit balance task to scheduler, if segment/channel's task already
exist in scheduler, submit task will failed.
to make TransferChannel/TransferSegment idempotent, we should skip to
submit if task already exist in scheduler.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: #36536
query coord use `segmentTaskDeleta/channelTaskDelta` to measure the
executing workload for querynode in scheduler, and we maintains the
`segmentTaskDeleta/channelTaskDelta` by `scheulder.Add(task)` and
`scheduler.remove(task)`, but `scheduler.remove(task)` has been called
in unexpected way, which cause a wrong
`segmentTaskDeleta/channelTaskDelta` value and affect the segment assign
logic, causes segment unbalance.
This PR moves to compute the `segmentTaskDeleta/channelTaskDelta` when
access, to avoid the wrong value affect.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Related to #36482
This PR reuses `SealedSegmentIDsRetrieved` field in `RetrieveResults`
struct to store segment id hint.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: https://github.com/milvus-io/milvus/issues/36182
* improved `Column.h` to make the code much more readable and
maintainable, and added detailed comments.
* fixed an issue where `ArrayColumn::NumRows()` always returns 0 when
the mmap backing storage is a file.
* removed unused `ColumnBase` constructors and unnecessary members so we
don't get confused.
* Updated `test_chunk_cache.cpp` to make the tests parameterized: to
test both mmap enabled and disabled. Added sparse field in the test to
add coverage.
* re-enabled test `Sealed::GetSparseVectorFromChunkCache`.
* But 2 other disabled tests `Sealed::WarmupChunkCache` and
`Sealed::GetVectorFromChunkCache` remain disabled, there seems to be
errors. @bigsheeper PTAL.
---------
Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
issue: #35821
After collection loaded, if we need to increase/decrease collection's
replica, we need to release and load it again.
milvus offers 4 solution to update loaded collection's replica, this PR
aims to dynamic change the replica number without release, and after
replica number changed, milvus will execute load replica or release
replica in async, and the replica loaded status can be checked by
getReplicas API.
Notice that if set too much replicas than querynode can afford，the new
replica won't be loaded successfully until enough querynode joins.
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: #36426
the old constriant requires only segment on current target can be
balanced, which is wrong, and caused that segment can't be move out from
stopping node, if it's only exist in next target.
by design, stopping balance need to move out all segment on it by
balance task, thus the unfair old constriant should be removed.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: #35859
This PR introduce two new param: toleranceFactor and checkRequestNum,
after every checkRequestNum request has been assigned, try to compute
querynode's workload score.
if the diff is less than the toleranceFactor, replica selection policy
will fallback to round_robin, which reduce the average cost to about
500ns.
if the diff is larger than the toleranceFactor, replica selection policy
will compute querynode's score to select the target node with smallest
score in every assigment.
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
#36252
remove no need type check. if users use null type writer to write
parquet, hope it successfully.
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
This PR introduce stats task for import:
1. Define new `Stats` and `IndexBuilding` states for importJob
2. Add new stats step to the import process: trigger the stats task and
wait for its completion
3. Abort stats task if import job failed
issue: https://github.com/milvus-io/milvus/issues/33744
---------
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
Flowgraph manager is not stopped durong datanode stopping procedure
which may lead to unexpect flowgraph behavior during/after datanode stop
progress.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #36257
during syncTargetVersion, sealed segment should be excluded, to avoid
it's growing segment be conusmed from stream again.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Related to #35996
For `Field Partial Load` feature, Milvus shall ignore index check for
non-loaded vector field.
Also, this PR unifies the logic of index check for load collection and
load partitions tasks.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #35303
This PR utilizes pk index in segment to exclude non-hit delete record
during load delete records. This ability is crucial when l0/delete
forward policy only replies on segment itself(without BF filtering).
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #36201
after querynode has been remove from replica, all dirty segment/channel
on it should be released.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Related to #36215
This PR add error handler setting option providing the possibility to
change error handling behavior other than panicking.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #35303
This PR add a param item to support change l0 forward behavior from bf
filtering and forward to remote load.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #35415
Currently ANN field is loaded in Parital load is not performed in proxy,
this cause error message not clear and no error returned when collection
is empty
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
See #36122
This PR is designed to enhance log performance through two improvements:
1. Optimize JSON encoding by switching JSON serializer to
`json-iterator`.
2. Adding support of lazy initialization `WithLazy`.
---------
Signed-off-by: Ted Xu <ted.xu@zilliz.com>
Related to #35941
Previous PR: #36034
This patch makes the switch branching logic correct and make the unit
test work for cases which does not select the whole dataset.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #36102
This PR use newly added `grpcSizeStatsHandler` to reduce calling
`proto.Size` since the request & response size info is recorded by grpc
framework.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #35941
Previous PR: #35943
This PR make `Trie` index using `MARISA_LABEL_ORDER`, which make
predictive search iterating in lexicographic order.
When trie index is build in label order, lexicographc could be utilized
accelerating `Range` operations.
However according to the official document, using `MARISA_LABEL_ORDER`
will make "exact match lookup, common prefix search, and predictive
search" slower.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #36064
when delegator has high memory usage, load l0 segment will failed. and
balance segment task will blocked by load segment task, then delegator
cann't free memory by moving out some segment, causes a logic dead lock.
this PR remove the limit for balance, we permit segment and balance
execute in parallel. which won't cause side effect due to:
1. one segment can only has one task in qc's scheduler, and load/release
task will replace balance task if necessary
2. balance speed has been limited, and it won't block load segment task.
3. if collection has load task and balance task at same time, load task
will be scheduled first due to high proirity.
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
fix not append valid data when transfer to insert record and add a tiny
check when in groupBy field.
#35924
Signed-off-by: lixinguo <xinguo.li@zilliz.com>
Co-authored-by: lixinguo <xinguo.li@zilliz.com>
issue: #33550
cause concurrent issue may occur between remove parition in target
manager and sync segment list to delegator. when it happens, some
segment may be released in delegator, and those segment may also be
synced to delegator, which cause delegator become unserviceable due to
lack of necessary segments, then search/query fails.
this PR make sure that all write access to target_manager will be
executed in serial to avoid the concurrent issues.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
This issue only occurs for a short time right after a table is created.
To avoid this, we simply reduce the frequency of cleaning up invalid
limiter nodes.
issue: https://github.com/milvus-io/milvus/issues/35933
---------
Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
Related to #35941
For marisa trie `predictive_search` default behavior, it value iterated
is not in lexicographic order.
This PR is a brute force fix to make range operator returns correct
values.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #35927
There are serveral issue this PR addresses:
- Use `ResetTraceConfig` method instead init one in update event handler
- Implement dynamic stats.Handler to receive tracing config update event
- Update `enable_trace` flag when `ResetTraceConfig` is invoked
- Change `enable_trace` to `std::atomic<bool>` in case of data race
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #35846
querycoord will notify proxy to update shard leader cache after
delegator location changes, but during querynode's failure recovery,
some delegator may become unserviceable due to lacking of segments, and
back to serviceable after segment loaded, so we also need to notify
proxy to invalidate shard leader cache when delegator serviceable state
changes.
This PR will maintain querynode's serviceable state during heartbeat,
and notify proxy to invalidate shard leader cache if serviceable state
changes.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Related to #35415
This PR make querycoord report error when load request tries to update
load fields list, which is currently not supported.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #33744
This PR includes the following changes:
1. Added a new task type to the task scheduler in datacoord: stats task,
which sorts segments by primary key.
2. Implemented segment sorting in indexnode.
3. Added a new field `FieldStatsLog` to SegmentInfo to store token index
information.
---------
Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
Bingyi Sun
Buqian Zheng
yihao.dai
congqixia
cqy123456
SimFG
smellthemoon
Zhen Ye
aoiasd
sre-ci-robot
Min Tian
jaime
XuanYang-cn
CharlesFeng
cai.zhang
zhagnlu
Ted Xu
Patrick Weizhi Xu
wayblink
wei liu
Rijin-N
Chun Han
sthuang
zhenshan.cao
Jiquan Long
OxalisCu
Abdullah Ahmed
foxspy