Previously, search with highlight only supported using BM25 search text
as the highlight target.
This PR adds support for highlighting with user-defined queries.
relate: https://github.com/milvus-io/milvus/issues/42589
---------
Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
issue: #44373
The current commit implements sparse filtering in query tasks using the
statistical information (Bloom filter/MinMax) of the Primary Key (PK).
The statistical information of the PK is bound to the segment during the
segment loading phase. A new filter has been added to the segment filter
to enable the sparse filtering functionality.
Signed-off-by: jiaqizho <jiaqi.zhou@zilliz.com>
issue: #42942
This pr includes the following changes:
1. Added checks for index checker in querycoord to generate drop index
tasks
2. Added drop index interface to querynode
3. To avoid search failure after dropping the index, the querynode
allows the use of lazy mode (warmup=disable) to load raw data even when
indexes contain raw data.
4. In segcore, loading the index no longer deletes raw data; instead, it
evicts it.
5. In expr, the index is pinned to prevent concurrent errors.
---------
Signed-off-by: sunby <sunbingyi1992@gmail.com>
issue: #41435
After introducing the caching layer's lazy loading and eviction
mechanisms, most parts of a segment won't be loaded into memory or disk
immediately, even if the segment is marked as LOADED. This means
physical resource usage may be very low. However, we still need to
reserve enough resources for the segments marked as LOADED. Thus, the
logic of resource usage estimation during segment loading, which based
on physcial resource usage only for now, should be changed.
To address this issue, we introduced the concept of logical resource
usage in this patch. This can be thought of as the base reserved
resource for each LOADED segment.
A segment’s logical resource usage is derived from its final evictable
and inevictable resource usage and calculated as follows:
```
SLR = SFPIER + evitable_cache_ratio * SFPER
```
it also equals to
```
SLR = (SFPIER + SFPER) - (1.0 - evitable_cache_ratio) * SFPER
```
`SLR`: The logical resource usage of a segment.
`SFPIER`: The final physical inevictable resource usage of a segment.
`SFPER`: The final physical evictable resource usage of a segment.
`evitable_cache_ratio`: The ratio of a segment's evictable resources
that can be cached locally. The higher the ratio, the more physical
memory is reserved for evictable memory.
When loading a segment, two types of resource usage are taken into
account.
First is the estimated maximum physical resource usage:
```
PPR = HPR + CPR + SMPR - SFPER
```
`PPR`: The predicted physical resource usage after the current segment
is allowed to load.
`HPR`: The physical resource usage obtained from hardware information.
`CPR`: The total physical resource usage of segments that have been
committed but not yet loaded. When one new segment is allow to load,
`CPR' = CPR + (SMR - SER)`. When one of the committed segments is
loaded, `CPR' = CPR - (SMR - SER)`.
`SMPR`: The maximum physical resource usage of the current segment.
`SFPER`: The final physical evictable resource usage of the current
segment.
Second is the estimated logical resource usage, this check is only valid
when eviction is enabled:
```
PLR = LLR + CLR + SLR
```
`PLR`: The predicted logical resource usage after the current segment is
allowed to load.
`LLR`: The total logical resource usage of all loaded segments. When a
new segment is loaded, `LLR` should be updated to `LLR' = LLR + SLR`.
`CLR`: The total logical resource usage of segments that have been
committed but not yet loaded. When one new segment is allow to load,
`CLR' = CLR + SLR`. When one of the committed segments is loaded, `CLR'
= CLR - SLR`.
`SLR`: The logical resource usage of the current segment.
Only when `PPR < PRL && PLR < PRL` (`PRL`: Physical resource limit of
the querynode), the segment is allowed to be loaded.
---------
Signed-off-by: Shawn Wang <shawn.wang@zilliz.com>
issue: #42884
Fixes an issue where delete records for a segment are lost from the
delete buffer if `load segment` execution on the delegator is too slow,
causing `syncTargetVersion` or other cleanup operations to clear them
prematurely.
Changes include:
- Introduced `Pin` and `Unpin` methods in `DeleteBuffer` interface and
its implementations (`doubleCacheBuffer`, `listDeleteBuffer`).
- Added a `pinnedTimestamps` map to track timestamps protected from
cleanup by specific segments.
- Modified `LoadSegments` in `shardDelegator` to `Pin` relevant segment
delete records before loading and `Unpin` them afterwards.
- Added `isPinned` check in `UnRegister` and `TryDiscard` methods of
`listDeleteBuffer` to skip cleanup if corresponding timestamps are
pinned.
- Added comprehensive unit tests for `Pin`, `Unpin`, and `isPinned`
functionality, covering basic, multiple pins, concurrent, and edge
cases.
This ensures the integrity of delete records by preventing their
premature removal from the delete buffer during segment loading.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Related to #43028
This PR:
- Add mutex prevent concurrent load segment & schema change
- Add schema verison field in load meta
- Update schema in PutOrRef if schema verison is larger
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #40532
- start timeticksync at rootcoord if the streaming service is not
available
- stop timeticksync if the streaming service is available
- open a read-only wal if some nodes in cluster is not upgrading to 2.6
- allow to open read-write wal after all nodes in cluster is upgrading
to 2.6
---------
Signed-off-by: chyezh <chyezh@outlook.com>
issue: https://github.com/milvus-io/milvus/issues/41690
This commit implements partial search result functionality when query
nodes go down, improving system availability during node failures. The
changes include:
- Enhanced load balancing in proxy (lb_policy.go) to handle node
failures with retry support
- Added partial search result capability in querynode delegator and
distribution logic
- Implemented tests for various partial result scenarios when nodes go
down
- Added metrics to track partial search results in querynode_metrics.go
- Updated parameter configuration to support partial result required
data ratio
- Replaced old partial_search_test.go with more comprehensive
partial_result_on_node_down_test.go
- Updated proto definitions and improved retry logic
These changes improve query resilience by returning partial results to
users when some query nodes are unavailable, ensuring that queries don't
completely fail when a portion of data remains accessible.
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: https://github.com/milvus-io/milvus/issues/41690
- Merge leader view and channel management into ChannelDistManager,
allowing a channel to have multiple delegators.
- Improve shard leader switching to ensure a single replica only has one
shard leader per channel. The shard leader handles all resource loading
and query requests.
- Refine the serviceable mechanism: after QC completes loading, sync the
query view to the delegator. The delegator then determines its
serviceable status based on the query view.
- When a delegator encounters forwarding query or deletion failures,
mark the corresponding segment as offline and transition it to an
unserviceable state.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: https://github.com/milvus-io/milvus/issues/41435
this PR is based on https://github.com/milvus-io/milvus/pull/41436.
Improvements include:
- Lazy Load support for Storage v1
- Use Low/High watermark to control eviction
- Caching Layer related config changes
- Removed ChunkCache related configs and code in golang
- Add `PinAllCells` helper method to CacheSlot class
- Modified ValueAt, RawAt, PrimitiveRawAt to Bulk version, to reduce
caching layer overhead
- Removed some unclear templated bulk_subscript methods
- CachedSearchIterator to store PinWrapper when searching on
ChunkedColumn, and removed unused contrustor.
---------
Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
issue: #38399
- add an adaptor type to adapt the streaming service client and
msgstream client to reuse the msgdispatcher.
Signed-off-by: chyezh <chyezh@outlook.com>
related to: #40884#39552
Reduce log frequency by aggregating deletion logs for L0 segments:
- Add segment count statistics in rangeHitL0Deletions function
- Change individual segment logs to a single consolidated log entry
- Include total number of processed L0 segments in log output
This change significantly reduces log volume while maintaining essential
visibility into deletion operations.
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue：#40207
related to https://github.com/milvus-io/milvus/pull/39552
- Correct comparison operator in UnRegister from > to >= to prevent
premature release of L0 segments with matching timestamps
- Add detailed logging for segment retention decisions during
unregistration
- Enhance error logging for buffer cleanup operations
- Add trace logs for segment registration/release lifecycle
- Include timestamp comparisons in debug logs for future troubleshooting
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
See also #40558
Related to #35303 & #38066 as well
This PR:
- Add `BufferedForward` to limit memory usage forwarding stream delete
- Add `UseLoad` flag to determine `Delete` shall use `segment.Delete` or
`segment.LoadDelta`
- Fix delegator accidentally use always true candidate while load
streaming delta
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #40292
related to #39552
- Fix incorrect delete checkpoint usage in SyncDistribution
- Change checkpoint parameter from action.GetCheckpoint() to
action.GetDeleteCP() in SyncTargetVersion call
- This resolves the issue where delete buffer data was being cleaned
prematurely due to wrong checkpoint reference
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: #39551
This PR remove querycoord's scheduling of l0 segments:
- only load l0 segment when watch channel
- only release l0 segment when release channel or sync data distribution
---------
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
issue: https://github.com/milvus-io/milvus/issues/39818
This PR mimics Varchar data type, allows insert, search, query, delete,
full-text search and others.
Functionalities related to filter expressions are disabled temporarily.
Storage changes for Text data type will be in the following PRs.
Signed-off-by: Patrick Weizhi Xu <weizhi.xu@zilliz.com>
The level zero mutex could be remove since all operations are guarded by
segment manager mutex
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
sparse vectors may have arbitrary number of non zeros and it is hard to
optimize without knowing the actual distribution of nnz. this PR adds a
metric for analyzing that.
issue: https://github.com/milvus-io/milvus/issues/35853
comparing with https://github.com/milvus-io/milvus/pull/38328, this
includes also metric for FTS in query node delegator
also fixed a bug of sparse when searching by pk
Signed-off-by: Buqian Zheng <zhengbuqian@gmail.com>
issue: #37908
cause paramtable is global single instance, which cause
paramtable.GetNodeID may return wrong server id in integration test.
This PR use node.GetNodeID to replace paramtable.GetNodeID
Signed-off-by: Wei Liu <wei.liu@zilliz.com>
Related to #35303
`deleteMut` shall be protecting streaming delete buffer, forward l0
could be move out of the rlock section to reduce tsafe impact from
loading segments.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #37177
Previous PR #37160
Collection meta is not ref-ed when loading l0 segment in `RemoteLoad`
policy, which cause collection meta release when lots of l0 segment
released.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #37112
Skip load logic used to work only when there is multiple segment load
info entires in load request. In continous delete case, delegator still
loads l0 segment, which occupies lot of memory.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #35303
Delta data is not needed when using `RemoteLoad` l0 forward policy. By
skipping load delta data, memory pressure could be eased if l0 segment
size/number is large.
Signed-off-by: Congqi Xia <congqi.xia@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>
