Add boltdb for volume needle map
boltdb is fairly slow to write, about 6 minutes for recreating index
for 1553934 files. Boltdb loads 1,553,934 x 16 = 24,862,944bytes from
disk, and generate the boltdb as large as 134,217,728 bytes in 6
minutes.
To compare, for leveldb, it recreates index in leveldb as large as
27,188,148 bytes in 8 seconds.
For in memory version, it loads the index in
To test the memory consumption, the leveldb or boltdb index are
created. And the server is restarted. Using the benchmark tool to read
lots of files. There are 7 volumes in benchmark collection, each with
about 1553K files.
For leveldb, the memory starts at 142,884KB, and stays at 179,340KB.
For boltdb, the memory starts at 73,756KB, and stays at 144,564KB.
For in-memory, the memory starts at 368,152KB, and stays at 448,032KB.
10 years ago Add boltdb for volume needle map
boltdb is fairly slow to write, about 6 minutes for recreating index
for 1553934 files. Boltdb loads 1,553,934 x 16 = 24,862,944bytes from
disk, and generate the boltdb as large as 134,217,728 bytes in 6
minutes.
To compare, for leveldb, it recreates index in leveldb as large as
27,188,148 bytes in 8 seconds.
For in memory version, it loads the index in
To test the memory consumption, the leveldb or boltdb index are
created. And the server is restarted. Using the benchmark tool to read
lots of files. There are 7 volumes in benchmark collection, each with
about 1553K files.
For leveldb, the memory starts at 142,884KB, and stays at 179,340KB.
For boltdb, the memory starts at 73,756KB, and stays at 144,564KB.
For in-memory, the memory starts at 368,152KB, and stays at 448,032KB.
10 years ago |
|
package storage
import ( "fmt" "os" "path/filepath"
"github.com/syndtr/goleveldb/leveldb/errors" "github.com/syndtr/goleveldb/leveldb/opt"
"github.com/chrislusf/seaweedfs/weed/storage/idx"
"github.com/syndtr/goleveldb/leveldb"
"github.com/chrislusf/seaweedfs/weed/glog" "github.com/chrislusf/seaweedfs/weed/storage/needle_map" . "github.com/chrislusf/seaweedfs/weed/storage/types" )
type LevelDbNeedleMap struct { baseNeedleMapper dbFileName string db *leveldb.DB }
func NewLevelDbNeedleMap(dbFileName string, indexFile *os.File, opts *opt.Options) (m *LevelDbNeedleMap, err error) { m = &LevelDbNeedleMap{dbFileName: dbFileName} m.indexFile = indexFile if !isLevelDbFresh(dbFileName, indexFile) { glog.V(1).Infof("Start to Generate %s from %s", dbFileName, indexFile.Name()) generateLevelDbFile(dbFileName, indexFile) glog.V(1).Infof("Finished Generating %s from %s", dbFileName, indexFile.Name()) } glog.V(1).Infof("Opening %s...", dbFileName)
if m.db, err = leveldb.OpenFile(dbFileName, opts); err != nil { if errors.IsCorrupted(err) { m.db, err = leveldb.RecoverFile(dbFileName, opts) } if err != nil { return } } glog.V(1).Infof("Loading %s...", indexFile.Name()) mm, indexLoadError := newNeedleMapMetricFromIndexFile(indexFile) if indexLoadError != nil { return nil, indexLoadError } m.mapMetric = *mm return }
func isLevelDbFresh(dbFileName string, indexFile *os.File) bool { // normally we always write to index file first
dbLogFile, err := os.Open(filepath.Join(dbFileName, "LOG")) if err != nil { return false } defer dbLogFile.Close() dbStat, dbStatErr := dbLogFile.Stat() indexStat, indexStatErr := indexFile.Stat() if dbStatErr != nil || indexStatErr != nil { glog.V(0).Infof("Can not stat file: %v and %v", dbStatErr, indexStatErr) return false }
return dbStat.ModTime().After(indexStat.ModTime()) }
func generateLevelDbFile(dbFileName string, indexFile *os.File) error { db, err := leveldb.OpenFile(dbFileName, nil) if err != nil { return err } defer db.Close() return idx.WalkIndexFile(indexFile, func(key NeedleId, offset Offset, size Size) error { if !offset.IsZero() && size.IsValid() { levelDbWrite(db, key, offset, size) } else { levelDbDelete(db, key) } return nil }) }
func (m *LevelDbNeedleMap) Get(key NeedleId) (element *needle_map.NeedleValue, ok bool) { bytes := make([]byte, NeedleIdSize) NeedleIdToBytes(bytes[0:NeedleIdSize], key) data, err := m.db.Get(bytes, nil) if err != nil || len(data) != OffsetSize+SizeSize { return nil, false } offset := BytesToOffset(data[0:OffsetSize]) size := BytesToSize(data[OffsetSize : OffsetSize+SizeSize]) return &needle_map.NeedleValue{Key: key, Offset: offset, Size: size}, true }
func (m *LevelDbNeedleMap) Put(key NeedleId, offset Offset, size Size) error { var oldSize Size if oldNeedle, ok := m.Get(key); ok { oldSize = oldNeedle.Size } m.logPut(key, oldSize, size) // write to index file first
if err := m.appendToIndexFile(key, offset, size); err != nil { return fmt.Errorf("cannot write to indexfile %s: %v", m.indexFile.Name(), err) } return levelDbWrite(m.db, key, offset, size) }
func levelDbWrite(db *leveldb.DB, key NeedleId, offset Offset, size Size) error {
bytes := needle_map.ToBytes(key, offset, size)
if err := db.Put(bytes[0:NeedleIdSize], bytes[NeedleIdSize:NeedleIdSize+OffsetSize+SizeSize], nil); err != nil { return fmt.Errorf("failed to write leveldb: %v", err) } return nil } func levelDbDelete(db *leveldb.DB, key NeedleId) error { bytes := make([]byte, NeedleIdSize) NeedleIdToBytes(bytes, key) return db.Delete(bytes, nil) }
func (m *LevelDbNeedleMap) Delete(key NeedleId, offset Offset) error { oldNeedle, found := m.Get(key) if !found || oldNeedle.Size.IsDeleted() { return nil } m.logDelete(oldNeedle.Size)
// write to index file first
if err := m.appendToIndexFile(key, offset, TombstoneFileSize); err != nil { return err }
return levelDbWrite(m.db, key, oldNeedle.Offset, -oldNeedle.Size) }
func (m *LevelDbNeedleMap) Close() { indexFileName := m.indexFile.Name() if err := m.indexFile.Sync(); err != nil { glog.Warningf("sync file %s failed: %v", indexFileName, err) } if err := m.indexFile.Close(); err != nil { glog.Warningf("close index file %s failed: %v", indexFileName, err) }
if err := m.db.Close(); err != nil { glog.Warningf("close levelDB failed: %v", err) } }
func (m *LevelDbNeedleMap) Destroy() error { m.Close() os.Remove(m.indexFile.Name()) return os.RemoveAll(m.dbFileName) }
|