milvus/internal/flushcommon/syncmgr/pack_writer_v2_test.go
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

270 lines
9.6 KiB
Go

// Licensed to the LF AI & Data foundation under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package syncmgr
import (
"context"
"fmt"
"math"
"math/rand"
"sync/atomic"
"testing"
"github.com/samber/lo"
"github.com/stretchr/testify/mock"
"github.com/stretchr/testify/suite"
"github.com/milvus-io/milvus-proto/go-api/v2/commonpb"
"github.com/milvus-io/milvus-proto/go-api/v2/schemapb"
"github.com/milvus-io/milvus/internal/allocator"
"github.com/milvus-io/milvus/internal/flushcommon/metacache"
"github.com/milvus-io/milvus/internal/flushcommon/metacache/pkoracle"
"github.com/milvus-io/milvus/internal/mocks/flushcommon/mock_util"
"github.com/milvus-io/milvus/internal/storage"
"github.com/milvus-io/milvus/internal/storagecommon"
"github.com/milvus-io/milvus/internal/storagev2/packed"
"github.com/milvus-io/milvus/internal/util/initcore"
"github.com/milvus-io/milvus/pkg/v2/common"
"github.com/milvus-io/milvus/pkg/v2/objectstorage"
"github.com/milvus-io/milvus/pkg/v2/proto/datapb"
"github.com/milvus-io/milvus/pkg/v2/util/paramtable"
"github.com/milvus-io/milvus/pkg/v2/util/typeutil"
)
func TestPackWriterV2Suite(t *testing.T) {
suite.Run(t, new(PackWriterV2Suite))
}
type PackWriterV2Suite struct {
suite.Suite
ctx context.Context
mockID atomic.Int64
logIDAlloc allocator.Interface
mockBinlogIO *mock_util.MockBinlogIO
schema *schemapb.CollectionSchema
cm storage.ChunkManager
rootPath string
maxRowNum int64
chunkSize uint64
currentSplit []storagecommon.ColumnGroup
}
func (s *PackWriterV2Suite) SetupTest() {
s.ctx = context.Background()
s.logIDAlloc = allocator.NewLocalAllocator(1, math.MaxInt64)
s.rootPath = "/tmp"
initcore.InitLocalArrowFileSystem(s.rootPath)
paramtable.Get().Init(paramtable.NewBaseTable())
s.schema = &schemapb.CollectionSchema{
Name: "sync_task_test_col",
Fields: []*schemapb.FieldSchema{
{FieldID: common.RowIDField, DataType: schemapb.DataType_Int64},
{FieldID: common.TimeStampField, DataType: schemapb.DataType_Int64},
{
FieldID: 100,
Name: "pk",
DataType: schemapb.DataType_Int64,
IsPrimaryKey: true,
},
{
FieldID: 101,
Name: "vector",
DataType: schemapb.DataType_FloatVector,
TypeParams: []*commonpb.KeyValuePair{
{Key: common.DimKey, Value: "128"},
},
},
},
StructArrayFields: []*schemapb.StructArrayFieldSchema{
{
FieldID: 102,
Name: "struct_array",
Fields: []*schemapb.FieldSchema{
{
FieldID: 103,
Name: "vector_array",
DataType: schemapb.DataType_ArrayOfVector,
ElementType: schemapb.DataType_FloatVector,
TypeParams: []*commonpb.KeyValuePair{
{Key: common.DimKey, Value: "128"},
},
},
},
},
},
}
allFields := typeutil.GetAllFieldSchemas(s.schema)
s.currentSplit = storagecommon.SplitColumns(allFields, map[int64]storagecommon.ColumnStats{}, storagecommon.NewSelectedDataTypePolicy(), storagecommon.NewRemanentShortPolicy(-1))
s.cm = storage.NewLocalChunkManager(objectstorage.RootPath(s.rootPath))
}
func (s *PackWriterV2Suite) TestPackWriterV2_Write() {
collectionID := int64(123)
partitionID := int64(456)
segmentID := int64(789)
channelName := fmt.Sprintf("by-dev-rootcoord-dml_0_%dv0", collectionID)
rows := 10
bfs := pkoracle.NewBloomFilterSet()
seg := metacache.NewSegmentInfo(&datapb.SegmentInfo{}, bfs, nil)
metacache.UpdateNumOfRows(1000)(seg)
mc := metacache.NewMockMetaCache(s.T())
mc.EXPECT().Collection().Return(collectionID).Maybe()
mc.EXPECT().GetSchema(mock.Anything).Return(s.schema).Maybe()
mc.EXPECT().GetSegmentByID(segmentID).Return(seg, true).Maybe()
mc.EXPECT().GetSegmentsBy(mock.Anything, mock.Anything).Return([]*metacache.SegmentInfo{seg}).Maybe()
mc.EXPECT().UpdateSegments(mock.Anything, mock.Anything).Run(func(action metacache.SegmentAction, filters ...metacache.SegmentFilter) {
action(seg)
}).Return().Maybe()
deletes := &storage.DeleteData{}
for i := 0; i < rows; i++ {
pk := storage.NewInt64PrimaryKey(int64(i + 1))
ts := uint64(100 + i)
deletes.Append(pk, ts)
}
pack := new(SyncPack).WithCollectionID(collectionID).WithPartitionID(partitionID).WithSegmentID(segmentID).WithChannelName(channelName).WithInsertData(genInsertData(rows, s.schema)).WithDeleteData(deletes)
bw := NewBulkPackWriterV2(mc, s.schema, s.cm, s.logIDAlloc, packed.DefaultWriteBufferSize, 0, nil, s.currentSplit)
gotInserts, _, _, _, _, _, err := bw.Write(context.Background(), pack)
s.NoError(err)
s.Equal(gotInserts[0].Binlogs[0].GetEntriesNum(), int64(rows))
s.Equal(gotInserts[0].Binlogs[0].GetLogPath(), "/tmp/insert_log/123/456/789/0/1")
s.Equal(gotInserts[101].Binlogs[0].GetLogPath(), "/tmp/insert_log/123/456/789/101/2")
}
func (s *PackWriterV2Suite) TestWriteEmptyInsertData() {
s.logIDAlloc = allocator.NewLocalAllocator(1, 1)
collectionID := int64(123)
partitionID := int64(456)
segmentID := int64(789)
channelName := fmt.Sprintf("by-dev-rootcoord-dml_0_%dv0", collectionID)
mc := metacache.NewMockMetaCache(s.T())
mc.EXPECT().GetSchema(mock.Anything).Return(s.schema).Maybe()
pack := new(SyncPack).WithCollectionID(collectionID).WithPartitionID(partitionID).WithSegmentID(segmentID).WithChannelName(channelName)
bw := NewBulkPackWriterV2(mc, s.schema, s.cm, s.logIDAlloc, packed.DefaultWriteBufferSize, 0, nil, s.currentSplit)
_, _, _, _, _, _, err := bw.Write(context.Background(), pack)
s.NoError(err)
}
func (s *PackWriterV2Suite) TestNoPkField() {
s.schema = &schemapb.CollectionSchema{
Name: "no pk field",
Fields: []*schemapb.FieldSchema{
{FieldID: common.RowIDField, DataType: schemapb.DataType_Int64},
{FieldID: common.TimeStampField, DataType: schemapb.DataType_Int64},
},
}
s.logIDAlloc = allocator.NewLocalAllocator(1, math.MaxInt64)
collectionID := int64(123)
partitionID := int64(456)
segmentID := int64(789)
channelName := fmt.Sprintf("by-dev-rootcoord-dml_0_%dv0", collectionID)
mc := metacache.NewMockMetaCache(s.T())
mc.EXPECT().GetSchema(mock.Anything).Return(s.schema).Maybe()
buf, _ := storage.NewInsertData(s.schema)
data := make(map[storage.FieldID]any)
data[common.RowIDField] = int64(1)
data[common.TimeStampField] = int64(1)
buf.Append(data)
pack := new(SyncPack).WithCollectionID(collectionID).WithPartitionID(partitionID).WithSegmentID(segmentID).WithChannelName(channelName).WithInsertData([]*storage.InsertData{buf})
bw := NewBulkPackWriterV2(mc, s.schema, s.cm, s.logIDAlloc, packed.DefaultWriteBufferSize, 0, nil, s.currentSplit)
_, _, _, _, _, _, err := bw.Write(context.Background(), pack)
s.Error(err)
}
func (s *PackWriterV2Suite) TestAllocIDExhausedError() {
s.logIDAlloc = allocator.NewLocalAllocator(1, 1)
collectionID := int64(123)
partitionID := int64(456)
segmentID := int64(789)
channelName := fmt.Sprintf("by-dev-rootcoord-dml_0_%dv0", collectionID)
rows := 10
mc := metacache.NewMockMetaCache(s.T())
mc.EXPECT().GetSchema(mock.Anything).Return(s.schema).Maybe()
pack := new(SyncPack).WithCollectionID(collectionID).WithPartitionID(partitionID).WithSegmentID(segmentID).WithChannelName(channelName).WithInsertData(genInsertData(rows, s.schema))
bw := NewBulkPackWriterV2(mc, s.schema, s.cm, s.logIDAlloc, packed.DefaultWriteBufferSize, 0, nil, s.currentSplit)
_, _, _, _, _, _, err := bw.Write(context.Background(), pack)
s.Error(err)
}
func (s *PackWriterV2Suite) TestWriteInsertDataError() {
s.logIDAlloc = allocator.NewLocalAllocator(1, math.MaxInt64)
collectionID := int64(123)
partitionID := int64(456)
segmentID := int64(789)
channelName := fmt.Sprintf("by-dev-rootcoord-dml_0_%dv0", collectionID)
mc := metacache.NewMockMetaCache(s.T())
mc.EXPECT().GetSchema(mock.Anything).Return(s.schema).Maybe()
buf, _ := storage.NewInsertData(s.schema)
data := make(map[storage.FieldID]any)
data[common.RowIDField] = int64(1)
buf.Append(data)
pack := new(SyncPack).WithCollectionID(collectionID).WithPartitionID(partitionID).WithSegmentID(segmentID).WithChannelName(channelName).WithInsertData([]*storage.InsertData{buf})
bw := NewBulkPackWriterV2(mc, s.schema, s.cm, s.logIDAlloc, packed.DefaultWriteBufferSize, 0, nil, s.currentSplit)
_, _, _, _, _, _, err := bw.Write(context.Background(), pack)
s.Error(err)
}
func genInsertData(size int, schema *schemapb.CollectionSchema) []*storage.InsertData {
buf, _ := storage.NewInsertData(schema)
for i := 0; i < size; i++ {
data := make(map[storage.FieldID]any)
data[common.RowIDField] = int64(i + 1)
data[common.TimeStampField] = int64(i + 1)
data[100] = int64(i + 1)
vector := lo.RepeatBy(128, func(_ int) float32 {
return rand.Float32()
})
data[101] = vector
arraySize := rand.Intn(3) + 2
vectorData := lo.RepeatBy(arraySize, func(_ int) float32 {
return rand.Float32()
})
vectorArray := &schemapb.VectorField{
Dim: 128,
Data: &schemapb.VectorField_FloatVector{
FloatVector: &schemapb.FloatArray{
Data: vectorData,
},
},
}
data[103] = vectorArray
buf.Append(data)
}
return []*storage.InsertData{buf}
}