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>
Related to #44956
Integrate the StorageV2 FFI interface as the unified storage layer for
reading packed columnar data, replacing the custom iterative reader with
a manifest-based approach using the milvus-storage library.
Changes:
- Add C++ FFI reader implementation (ffi_reader_c.cpp/h) with Arrow C
Stream interface
- Implement utility functions to convert CStorageConfig to
milvus-storage Properties
- Create ManifestReader in Go that generates manifests from binlogs
- Add FFI packed reader CGO bindings (packed_reader_ffi.go)
- Refactor NewBinlogRecordReader to use ManifestReader for V2 storage
- Support both manifest file paths and direct manifest content
- Enable configurable buffer sizes and column projection
Technical improvements:
- Zero-copy data exchange using Arrow C Data Interface
- Optimized I/O operations through milvus-storage library
- Simplified code path with manifest-based reading
- Better performance with batched streaming reads
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
See #39173
In this PR:
- Adjusted the delta log serialization APIs.
- Refactored the stats collector to improve the collection and digest of
primary key and BM25 statistics.
- Introduced new tests for the delta log reader/writer and stats
collectors to ensure functionality and correctness.
---------
Signed-off-by: Ted Xu <ted.xu@zilliz.com>
issue: https://github.com/milvus-io/milvus/issues/42148
Optimized from
Go VectorArray → VectorArray Proto → Binary → C++ VectorArray Proto →
C++ VectorArray local impl → Memory
to
Go VectorArray → Arrow ListArray → Memory
---------
Signed-off-by: SpadeA <tangchenjie1210@gmail.com>
1. Enable Milvus to read cipher configs
2. Enable cipher plugin in binlog reader and writer
3. Add a testCipher for unittests
4. Support pooling for datanode
5. Add encryption in storagev2
See also: #40321
Signed-off-by: yangxuan <xuan.yang@zilliz.com>
---------
Signed-off-by: yangxuan <xuan.yang@zilliz.com>
Ref https://github.com/milvus-io/milvus/issues/42148
This PR supports create index for vector array (now, only for
`DataType.FLOAT_VECTOR`) and search on it.
The index type supported in this PR is `EMB_LIST_HNSW` and the metric
type is `MAX_SIM` only.
The way to use it:
```python
milvus_client = MilvusClient("xxx:19530")
schema = milvus_client.create_schema(enable_dynamic_field=True, auto_id=True)
...
struct_schema = milvus_client.create_struct_array_field_schema("struct_array_field")
...
struct_schema.add_field("struct_float_vec", DataType.ARRAY_OF_VECTOR, element_type=DataType.FLOAT_VECTOR, dim=128, max_capacity=1000)
...
schema.add_struct_array_field(struct_schema)
index_params = milvus_client.prepare_index_params()
index_params.add_index(field_name="struct_float_vec", index_type="EMB_LIST_HNSW", metric_type="MAX_SIM", index_params={"nlist": 128})
...
milvus_client.create_index(COLLECTION_NAME, schema=schema, index_params=index_params)
```
Note: This PR uses `Lims` to convey offsets of the vector array to
knowhere where vectors of multiple vector arrays are concatenated and we
need offsets to specify which vectors belong to which vector array.
---------
Signed-off-by: SpadeA <tangchenjie1210@gmail.com>
Signed-off-by: SpadeA-Tang <tangchenjie1210@gmail.com>
Add `arrowBuild.Reserve` call for `ValueSerializer` to reduce repeated
resizing buffer when write size is large
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Ref https://github.com/milvus-io/milvus/issues/42148https://github.com/milvus-io/milvus/pull/42406 impls the segcore part of
storage for handling with VectorArray.
This PR:
1. impls the go part of storage for VectorArray
2. impls the collection creation with StructArrayField and VectorArray
3. insert and retrieve data from the collection.
---------
Signed-off-by: SpadeA <tangchenjie1210@gmail.com>
Signed-off-by: SpadeA-Tang <tangchenjie1210@gmail.com>
Signed-off-by: SpadeA-Tang <u6748471@anu.edu.au>
This parameter determines whether the returned value should be a copy or
a reference from the arrow array. The updates enhance memory management
and provide more control over data handling during deserialization.
See #43186
---------
Signed-off-by: Ted Xu <ted.xu@zilliz.com>
Related to #43522
Currently, passing partial schema to storage v2 packed reader may
trigger SEGV during clustering compaction unit test.
This patch implement `NeededFields` differently in each `RecordReader`
imlementation. For now, v2 will implemented as no-op. This will be
supported after packed reader support this API.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
Related to #42856
Default value will be missing after segment get sorted/compacted. This
PR is a temp workaround since in long term default value shall be filled
with storage engine instead.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
- Feat: Support Mix compaction. Covering tests include compatibility and
rollback ability.
- Read v1 segments and compact with v2 format.
- Read both v1 and v2 segments and compact with v2 format.
- Read v2 segments and compact with v2 format.
- Compact with duplicate primary key test.
- Compact with bm25 segments.
- Compact with merge sort segments.
- Compact with no expiration segments.
- Compact with lack binlog segments.
- Compact with nullable field segments.
- Feat: Support Clustering compaction. Covering tests include
compatibility and rollback ability.
- Read v1 segments and compact with v2 format.
- Read both v1 and v2 segments and compact with v2 format.
- Read v2 segments and compact with v2 format.
- Compact bm25 segments with v2 format.
- Compact with memory limit.
- Enhance: Use serdeMap serialize in BuildRecord function to support all
Milvus data types.
related: #39173
Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
* use the new packed reader and writer api to be compatible with current
etcd meta
* For the new packed writer API: column groups and paths are explicitly
defined by users and won't split column groups by memory in storage v2.
Packed writer follows the user-defined column groups to split arrow
record and write into the corresponding file path.
* For the new packed reader API: read paths are explicitly defined by
users.
related: #39173
Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
issue: #34357
Go Parquet uses dictionary encoding by default, and it will fall back to
plain encoding if the dictionary size exceeds the dictionary size page
limit. Users can specify custom fallback encoding by using
`parquet.WithEncoding(ENCODING_METHOD)` in writer properties. However,
Go Parquet [fallbacks to plain
encoding](e65c1e295d/go/parquet/file/column_writer_types.gen.go.tmpl (L238))
rather than custom encoding method users provide. Therefore, this patch
only turns off dictionary encoding for the primary key.
With a 5 million auto ID primary key benchmark, the parquet file size
improves from 13.93 MB to 8.36 MB when dictionary encoding is turned
off, reducing primary key storage space by 40%.
Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
See also #34483
Some lint issues are introduced due to lack of static check run. This PR
fixes these problems.
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
issue: #34123
Benchmark case: The benchmark run the go benchmark function
`BenchmarkDeltalogFormat` which is put in the Files changed. It tests
the performance of serializing and deserializing from two different data
formats under a 10 million delete log dataset.
Metrics: The benchmarks measure the average time taken per operation
(ns/op), memory allocated per operation (MB/op), and the number of
memory allocations per operation (allocs/op).
| Test Name | Avg Time (ns/op) | Time Comparison | Memory Allocation
(MB/op) | Memory Comparison | Allocation Count (allocs/op) | Allocation
Comparison |
|---------------------------------|------------------|-----------------|---------------------------|-------------------|------------------------------|------------------------|
| one_string_format_reader | 2,781,990,000 | Baseline | 2,422 | Baseline
| 20,336,539 | Baseline |
| pk_ts_separate_format_reader | 480,682,639 | -82.72% | 1,765 | -27.14%
| 20,396,958 | +0.30% |
| one_string_format_writer | 5,483,436,041 | Baseline | 13,900 |
Baseline | 70,057,473 | Baseline |
| pk_and_ts_separate_format_writer| 798,591,584 | -85.43% | 2,178 |
-84.34% | 30,270,488 | -56.78% |
Both read and write operations show significant improvements in both
speed and memory allocation.
Signed-off-by: shaoting-huang <shaoting.huang@zilliz.com>
See also #33787
The parsing delete log is distributed in lots of places, which is not
recommended and hard to maintain.
This PR abstract common parsing logic into `DeleteLog.Parse` method to
unify implementation and make it easier to replace json parsing lib.
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
---------
Signed-off-by: Congqi Xia <congqi.xia@zilliz.com>
