27 Commits

Author SHA1 Message Date
congqixia
c01fd94a6a
enhance: integrate Storage V2 FFI interface for unified storage access (#45723)
Related #44956
This commit integrates the Storage V2 FFI (Foreign Function Interface)
interface throughout the Milvus codebase, enabling unified storage
access through the Loon FFI layer. This is a significant step towards
standardizing storage operations across different storage versions.

1. Configuration Support
- **configs/milvus.yaml**: Added `useLoonFFI` configuration flag under
`common.storage.file.splitByAvgSize` section
- Allows runtime toggle between traditional binlog readers and new
FFI-based manifest readers
  - Default: `false` (maintains backward compatibility)

2. Core FFI Infrastructure

Enhanced Utilities (internal/core/src/storage/loon_ffi/util.cpp/h)
- **ToCStorageConfig()**: Converts Go's `StorageConfig` to C's
`CStorageConfig` struct for FFI calls
- **GetManifest()**: Parses manifest JSON and retrieves latest column
groups using FFI
  - Accepts manifest path with `base_path` and `ver` fields
  - Calls `get_latest_column_groups()` FFI function
  - Returns column group information as string
  - Comprehensive error handling for JSON parsing and FFI errors

3. Dependency Updates
- **internal/core/thirdparty/milvus-storage/CMakeLists.txt**:
  - Updated milvus-storage version from `0883026` to `302143c`
  - Ensures compatibility with latest FFI interfaces

4. Data Coordinator Changes

All compaction task builders now include manifest path in segment
binlogs:

- **compaction_task_clustering.go**: Added `Manifest:
segInfo.GetManifestPath()` to segment binlogs
- **compaction_task_l0.go**: Added manifest path to both L0 segment
selection and compaction plan building
- **compaction_task_mix.go**: Added manifest path to mixed compaction
segment binlogs
- **meta.go**: Updated metadata completion logic:
- `completeClusterCompactionMutation()`: Set `ManifestPath` in new
segment info
- `completeMixCompactionMutation()`: Preserve manifest path in compacted
segments
- `completeSortCompactionMutation()`: Include manifest path in sorted
segments

5. Data Node Compactor Enhancements

All compactors updated to support dual-mode reading (binlog vs
manifest):

6. Flush & Sync Manager Updates

Pack Writer V2 (pack_writer_v2.go)
- **BulkPackWriterV2.Write()**: Extended return signature to include
`manifest string`
- Implementation:
  - Generate manifest path: `path.Join(pack.segmentID, "manifest.json")`
  - Write packed data using FFI-based writer
  - Return manifest path along with binlogs, deltas, and stats

Task Handling (task.go)
- Updated all sync task result handling to accommodate new manifest
return value
- Ensured backward compatibility for callers not using manifest

7. Go Storage Layer Integration

New Interfaces and Implementations
- **record_reader.go**: Interface for unified record reading across
storage versions
- **record_writer.go**: Interface for unified record writing across
storage versions
- **binlog_record_writer.go**: Concrete implementation for traditional
binlog-based writing

Enhanced Schema Support (schema.go, schema_test.go)
- Schema conversion utilities to support FFI-based storage operations
- Ensures proper Arrow schema mapping for V2 storage

Serialization Updates
- **serde.go, serde_events.go, serde_events_v2.go**: Updated to work
with new reader/writer interfaces
- Test files updated to validate dual-mode serialization

8. Storage V2 Packed Format

FFI Common (storagev2/packed/ffi_common.go)
- Common FFI utilities and type conversions for packed storage format

Packed Writer FFI (storagev2/packed/packed_writer_ffi.go)
- FFI-based implementation of packed writer
- Integrates with Loon storage layer for efficient columnar writes

Packed Reader FFI (storagev2/packed/packed_reader_ffi.go)
- Already existed, now complemented by writer implementation

9. Protocol Buffer Updates

data_coord.proto & datapb/data_coord.pb.go
- Added `manifest` field to compaction segment messages
- Enables passing manifest metadata through compaction pipeline

worker.proto & workerpb/worker.pb.go
- Added compaction parameter for `useLoonFFI` flag
- Allows workers to receive FFI configuration from coordinator

10. Parameter Configuration

component_param.go
- Added `UseLoonFFI` parameter to compaction configuration
- Reads from `common.storage.file.useLoonFFI` config path
- Default: `false` for safe rollout

11. Test Updates
- **clustering_compactor_storage_v2_test.go**: Updated signatures to
handle manifest return value
- **mix_compactor_storage_v2_test.go**: Updated test helpers for
manifest support
- **namespace_compactor_test.go**: Adjusted writer calls to expect
manifest
- **pack_writer_v2_test.go**: Validated manifest generation in pack
writing

This integration follows a **dual-mode approach**:
1. **Legacy Path**: Traditional binlog-based reading/writing (when
`useLoonFFI=false` or no manifest)
2. **FFI Path**: Manifest-based reading/writing through Loon FFI (when
`useLoonFFI=true` and manifest exists)

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-11-24 19:57:07 +08:00
congqixia
569a5b40d2
enhance: [StorageV2] add manifest path support for FFI integration (#44991)
Related to #44956

Add manifest_path field throughout the data path to support LOON Storage
V2 manifest tracking. The manifest stores metadata for segment data
files and enables the unified Storage V2 FFI interface.

Changes include:
- Add manifest_path field to SegmentInfo and SaveBinlogPathsRequest
proto messages
- Add UpdateManifest operator to datacoord meta operations
- Update metacache, sync manager, and meta writer to propagate manifest
paths
- Include manifest_path in segment load info for query coordinator

This is part of the Storage V2 FFI interface integration.

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-10-27 19:24:10 +08:00
congqixia
1e3ec42e54
fix: Check child fields len instead of nil (#44405)
Related to #44398

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-09-17 11:36:00 +08:00
congqixia
fc968ff1c2
enhance: [StorageV2] Pass args for avg size split policy (#44301)
Related to #44257

This PR
- Pass column stats for avg size split policy
- Add param items for policy configuration

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-09-11 10:43:57 +08:00
congqixia
f5618d5153
enhance: [StorageV2] Utilized advance split policy and persist in meta (#44282)
Related to #44257

This PR:
- Utilize configurable split policy for storage v2, enabling system
field policy
- Store split result in field binlog struct
- Adapt legacy binlog without child fields

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-09-10 14:47:57 +08:00
sthuang
a2c7ed2780
fix: [StorageV2] sort field binlogs paths for packed reader and writer (#43585)
key changes:
* fix unstable storage v2 compaction unit test by guaranteeing the order
of paths during sync.
* bump milvus-storage version, include
https://github.com/milvus-io/milvus-storage/pull/222
https://github.com/milvus-io/milvus-storage/pull/223
https://github.com/milvus-io/milvus-storage/pull/224
https://github.com/milvus-io/milvus-storage/pull/225
https://github.com/milvus-io/milvus-storage/pull/226
* Also fix the below related oom issue.
related: https://github.com/milvus-io/milvus/issues/43310

Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
2025-07-30 08:09:36 +08:00
Zhen Ye
25b76e1fde
fix: cannot auto balance the channel from old arch to streamingnode (#43424)
issue: #43416, #43413

- also fix the panic on streamingnode when concurrent sync

Signed-off-by: chyezh <chyezh@outlook.com>
2025-07-20 23:00:52 +08:00
cai.zhang
3ffd44f302
fix: Fix remaining issues with Datanode pooling and StorageV2 (#43147)
issue: #43146

---------

Signed-off-by: Cai Zhang <cai.zhang@zilliz.com>
2025-07-10 14:26:48 +08:00
yihao.dai
ed55b14484
fix: Release data memory after sync task completes (#42627)
Release data memory after sync task completes to prevent datanode oom
during import.

issue: https://github.com/milvus-io/milvus/issues/42608

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2025-06-10 16:28:34 +08:00
congqixia
b8d7045539
enhance: [Add Field] Use consistent schema for single buffer (#41891)
Related to #41873

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-05-17 19:46:22 +08:00
XuanYang-cn
e6c46a25ea
enhance: Use correct counter metrics for overall wa calculation (#40394)
- Use CounterVec to calculate sum of increase during a time period.
- Use entries number instead of binlog size

---------

Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2025-03-10 16:34:06 +08:00
sthuang
63a7c4570e
feat: storage v2 sync (#39663)
related: #39173

Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
2025-03-05 11:22:15 +08:00
congqixia
cb7f2fa6fd
enhance: Use v2 package name for pkg module (#39990)
Related to #39095

https://go.dev/doc/modules/version-numbers

Update pkg version according to golang dep version convention

---------

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2025-02-22 23:15:58 +08:00
XuanYang-cn
1f14053c70
enhance: Enable to observe write amplification (#39661)
Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2025-02-08 18:38:43 +08:00
Ted Xu
56659bacbb
enhance: make serialization be part of sync task to support file format change (#38946)
See #38945

---------

Signed-off-by: Ted Xu <ted.xu@zilliz.com>
2025-01-23 15:49:05 +08:00
aoiasd
9cb4c4e8ac
fix: bm25 import segment without bm25 stats meta (#38855)
relate: https://github.com/milvus-io/milvus/issues/38854

Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2025-01-21 11:09:04 +08:00
Zhen Ye
bb8d1ab3bf
enhance: make new go package to manage proto (#39114)
issue: #39095

---------

Signed-off-by: chyezh <chyezh@outlook.com>
2025-01-10 10:49:01 +08:00
congqixia
b0bd290a6e
enhance: Use internal json(sonic) to replace std json lib (#37708)
Related to #35020

Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
2024-11-18 10:46:31 +08:00
XuanYang-cn
5a23c80f20
fix: Change memoryCheck write lock to read lock (#37525)
See also: milvus-io#37493

Signed-off-by: yangxuan <xuan.yang@zilliz.com>

Signed-off-by: yangxuan <xuan.yang@zilliz.com>
2024-11-15 10:44:31 +08:00
jaime
1d06d4324b
fix: Int64 overflow in JSON encoding (#37657)
issue: ##36621

- For simple types in a struct, add "string" to the JSON tag for
automatic string conversion during JSON encoding.
- For complex types in a struct, replace "int64" with "string."

Signed-off-by: jaime <yun.zhang@zilliz.com>
2024-11-14 22:52:30 +08:00
jaime
f348bd9441
feat: add segment,pipeline, replica and resourcegroup api for WebUI (#37344)
issue: #36621

Signed-off-by: jaime <yun.zhang@zilliz.com>
2024-11-07 11:52:25 +08:00
jaime
9d16b972ea
feat: add tasks page into management WebUI (#37002)
issue: #36621

1. Add API to access task runtime metrics, including:
  - build index task
  - compaction task
  - import task
- balance (including load/release of segments/channels and some leader
tasks on querycoord)
  - sync task
2. Add a debug model to the webpage by using debug=true or debug=false
in the URL query parameters to enable or disable debug mode.

Signed-off-by: jaime <yun.zhang@zilliz.com>
2024-10-28 10:13:29 +08:00
aoiasd
20750c061b
fix: loss small bm25 binlogs (#36942)
Sync task don't flush small bm25 logs, cause growing segment bm25 stats
loss.
relate: https://github.com/milvus-io/milvus/issues/36805

Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-10-17 14:49:23 +08:00
yihao.dai
80f25d497f
enhance: Add metrics to monitor import throughput and imported rows (#36519)
issue: https://github.com/milvus-io/milvus/issues/36518

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-09-28 17:31:15 +08:00
aoiasd
139787371e
feat: support embedding bm25 sparse vector and flush bm25 stats log (#36036)
relate: https://github.com/milvus-io/milvus/issues/35853

---------

Signed-off-by: aoiasd <zhicheng.yue@zilliz.com>
2024-09-19 10:57:12 +08:00
yihao.dai
6130a85444
enhance: Remove bf from streaming node (#35902)
Remove bf from streaming node:
1. When watching vchannels, skip loading bloom filters for segments.
2. Bypass bloom filter checks for delete messages, directly writing to
L0 segments.
3. Remove flushed segments proactively after flush.

issue: https://github.com/milvus-io/milvus/issues/33285,
https://github.com/milvus-io/milvus/issues/34585

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-09-03 14:17:02 +08:00
yihao.dai
8aab6cbfac
enhance: Organize the common modules of streamingNode and dataNode (#34773)
1. Move the common modules of streamingNode and dataNode to flushcommon
2. Add new GetVChannels interface for rootcoord

issue: https://github.com/milvus-io/milvus/issues/33285

---------

Signed-off-by: bigsheeper <yihao.dai@zilliz.com>
2024-07-22 11:33:51 +08:00