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.
2015-03-29 11:04:32 -07:00
parent 0d23f49f74
commit 1b6ab2f6af
13 changed files with 310 additions and 89 deletions
--- a/go/weed/compact.go
+++ b/go/weed/compact.go
@@ -33,7 +33,8 @@ func runCompact(cmd *Command, args []string) bool {
 	}

 	vid := storage.VolumeId(*compactVolumeId)
-	v, err := storage.NewVolume(*compactVolumePath, *compactVolumeCollection, vid, false, nil, nil)
+	v, err := storage.NewVolume(*compactVolumePath, *compactVolumeCollection, vid,
+		storage.NeedleMapInMemory, nil, nil)
 	if err != nil {
 		glog.Fatalf("Load Volume [ERROR] %s\n", err)
 	}
--- a/go/weed/export.go
+++ b/go/weed/export.go
@@ -99,23 +99,25 @@ func runExport(cmd *Command, args []string) bool {

 	var version storage.Version

-	err = storage.ScanVolumeFile(*exportVolumePath, *exportCollection, vid, func(superBlock storage.SuperBlock) error {
-		version = superBlock.Version()
-		return nil
-	}, true, func(n *storage.Needle, offset int64) error {
-		nv, ok := nm.Get(n.Id)
-		glog.V(3).Infof("key %d offset %d size %d disk_size %d gzip %v ok %v nv %+v",
-			n.Id, offset, n.Size, n.DiskSize(), n.IsGzipped(), ok, nv)
-		if ok && nv.Size > 0 && int64(nv.Offset)*8 == offset {
-			return walker(vid, n, version)
-		}
-		if !ok {
-			glog.V(2).Infof("This seems deleted %d size %d", n.Id, n.Size)
-		} else {
-			glog.V(2).Infof("Skipping later-updated Id %d size %d", n.Id, n.Size)
-		}
-		return nil
-	})
+	err = storage.ScanVolumeFile(*exportVolumePath, *exportCollection, vid,
+		storage.NeedleMapInMemory,
+		func(superBlock storage.SuperBlock) error {
+			version = superBlock.Version()
+			return nil
+		}, true, func(n *storage.Needle, offset int64) error {
+			nv, ok := nm.Get(n.Id)
+			glog.V(3).Infof("key %d offset %d size %d disk_size %d gzip %v ok %v nv %+v",
+				n.Id, offset, n.Size, n.DiskSize(), n.IsGzipped(), ok, nv)
+			if ok && nv.Size > 0 && int64(nv.Offset)*8 == offset {
+				return walker(vid, n, version)
+			}
+			if !ok {
+				glog.V(2).Infof("This seems deleted %d size %d", n.Id, n.Size)
+			} else {
+				glog.V(2).Infof("Skipping later-updated Id %d size %d", n.Id, n.Size)
+			}
+			return nil
+		})
 	if err != nil {
 		glog.Fatalf("Export Volume File [ERROR] %s\n", err)
 	}
--- a/go/weed/fix.go
+++ b/go/weed/fix.go
@@ -47,19 +47,21 @@ func runFix(cmd *Command, args []string) bool {
 	defer nm.Close()

 	vid := storage.VolumeId(*fixVolumeId)
-	err = storage.ScanVolumeFile(*fixVolumePath, *fixVolumeCollection, vid, func(superBlock storage.SuperBlock) error {
-		return nil
-	}, false, func(n *storage.Needle, offset int64) error {
-		glog.V(2).Infof("key %d offset %d size %d disk_size %d gzip %v", n.Id, offset, n.Size, n.DiskSize(), n.IsGzipped())
-		if n.Size > 0 {
-			pe := nm.Put(n.Id, uint32(offset/storage.NeedlePaddingSize), n.Size)
-			glog.V(2).Infof("saved %d with error %v", n.Size, pe)
-		} else {
-			glog.V(2).Infof("skipping deleted file ...")
-			return nm.Delete(n.Id)
-		}
-		return nil
-	})
+	err = storage.ScanVolumeFile(*fixVolumePath, *fixVolumeCollection, vid,
+		storage.NeedleMapInMemory,
+		func(superBlock storage.SuperBlock) error {
+			return nil
+		}, false, func(n *storage.Needle, offset int64) error {
+			glog.V(2).Infof("key %d offset %d size %d disk_size %d gzip %v", n.Id, offset, n.Size, n.DiskSize(), n.IsGzipped())
+			if n.Size > 0 {
+				pe := nm.Put(n.Id, uint32(offset/storage.NeedlePaddingSize), n.Size)
+				glog.V(2).Infof("saved %d with error %v", n.Size, pe)
+			} else {
+				glog.V(2).Infof("skipping deleted file ...")
+				return nm.Delete(n.Id)
+			}
+			return nil
+		})
 	if err != nil {
 		glog.Fatalf("Export Volume File [ERROR] %s\n", err)
 	}
--- a/go/weed/server.go
+++ b/go/weed/server.go
@@ -11,6 +11,7 @@ import (
 	"time"

 	"github.com/chrislusf/weed-fs/go/glog"
+	"github.com/chrislusf/weed-fs/go/storage"
 	"github.com/chrislusf/weed-fs/go/util"
 	"github.com/chrislusf/weed-fs/go/weed/weed_server"
 	"github.com/gorilla/mux"
@@ -68,7 +69,7 @@ var (
 	volumeDataFolders             = cmdServer.Flag.String("dir", os.TempDir(), "directories to store data files. dir[,dir]...")
 	volumeMaxDataVolumeCounts     = cmdServer.Flag.String("volume.max", "7", "maximum numbers of volumes, count[,count]...")
 	volumePulse                   = cmdServer.Flag.Int("pulseSeconds", 5, "number of seconds between heartbeats")
-	volumeUseLevelDb              = cmdServer.Flag.Bool("volume.leveldb", false, "Change to leveldb mode to save memory with reduced performance of read and write.")
+	volumeIndexType               = cmdServer.Flag.String("volume.index", "memory", "Choose [memory|leveldb|boltdb] mode for memory~performance balance.")
 	volumeFixJpgOrientation       = cmdServer.Flag.Bool("volume.images.fix.orientation", true, "Adjust jpg orientation when uploading.")
 	isStartingFiler               = cmdServer.Flag.Bool("filer", false, "whether to start filer")

@@ -233,10 +234,17 @@ func runServer(cmd *Command, args []string) bool {
 	if isSeperatedPublicPort {
 		publicVolumeMux = http.NewServeMux()
 	}
+	volumeNeedleMapKind := storage.NeedleMapInMemory
+	switch *volumeIndexType {
+	case "leveldb":
+		volumeNeedleMapKind = storage.NeedleMapLevelDb
+	case "boltdb":
+		volumeNeedleMapKind = storage.NeedleMapBoltDb
+	}
 	volumeServer := weed_server.NewVolumeServer(volumeMux, publicVolumeMux,
 		*serverIp, *volumePort, *serverPublicUrl,
 		folders, maxCounts,
-		*volumeUseLevelDb,
+		volumeNeedleMapKind,
 		*serverIp+":"+strconv.Itoa(*masterPort), *volumePulse, *serverDataCenter, *serverRack,
 		serverWhiteList, *volumeFixJpgOrientation,
 	)
--- a/go/weed/volume.go
+++ b/go/weed/volume.go
@@ -9,6 +9,7 @@ import (
 	"time"

 	"github.com/chrislusf/weed-fs/go/glog"
+	"github.com/chrislusf/weed-fs/go/storage"
 	"github.com/chrislusf/weed-fs/go/util"
 	"github.com/chrislusf/weed-fs/go/weed/weed_server"
 )
@@ -32,7 +33,7 @@ type VolumeServerOptions struct {
 	dataCenter            *string
 	rack                  *string
 	whiteList             []string
-	useLevelDb            *bool
+	indexType             *string
 	fixJpgOrientation     *bool
 }

@@ -49,7 +50,7 @@ func init() {
 	v.maxCpu = cmdVolume.Flag.Int("maxCpu", 0, "maximum number of CPUs. 0 means all available CPUs")
 	v.dataCenter = cmdVolume.Flag.String("dataCenter", "", "current volume server's data center name")
 	v.rack = cmdVolume.Flag.String("rack", "", "current volume server's rack name")
-	v.useLevelDb = cmdVolume.Flag.Bool("leveldb", false, "Change to leveldb mode to save memory with reduced performance of read and write.")
+	v.indexType = cmdVolume.Flag.String("index", "memory", "Choose [memory|leveldb|boltdb] mode for memory~performance balance.")
 	v.fixJpgOrientation = cmdVolume.Flag.Bool("images.fix.orientation", true, "Adjust jpg orientation when uploading.")
 }

@@ -115,10 +116,17 @@ func runVolume(cmd *Command, args []string) bool {
 		publicVolumeMux = http.NewServeMux()
 	}

+	volumeNeedleMapKind := storage.NeedleMapInMemory
+	switch *v.indexType {
+	case "leveldb":
+		volumeNeedleMapKind = storage.NeedleMapLevelDb
+	case "boltdb":
+		volumeNeedleMapKind = storage.NeedleMapBoltDb
+	}
 	volumeServer := weed_server.NewVolumeServer(volumeMux, publicVolumeMux,
 		*v.ip, *v.port, *v.publicUrl,
 		v.folders, v.folderMaxLimits,
-		*v.useLevelDb,
+		volumeNeedleMapKind,
 		*v.master, *v.pulseSeconds, *v.dataCenter, *v.rack,
 		v.whiteList,
 		*v.fixJpgOrientation,
--- a/go/weed/weed_server/volume_server.go
+++ b/go/weed/weed_server/volume_server.go
@@ -20,14 +20,14 @@ type VolumeServer struct {
 	store        *storage.Store
 	guard        *security.Guard

-	UseLevelDb        bool
+	needleMapKind     storage.NeedleMapType
 	FixJpgOrientation bool
 }

 func NewVolumeServer(adminMux, publicMux *http.ServeMux, ip string,
 	port int, publicUrl string,
 	folders []string, maxCounts []int,
-	useLevelDb bool,
+	needleMapKind storage.NeedleMapType,
 	masterNode string, pulseSeconds int,
 	dataCenter string, rack string,
 	whiteList []string,
@@ -36,11 +36,11 @@ func NewVolumeServer(adminMux, publicMux *http.ServeMux, ip string,
 		pulseSeconds:      pulseSeconds,
 		dataCenter:        dataCenter,
 		rack:              rack,
-		UseLevelDb:        useLevelDb,
+		needleMapKind:     needleMapKind,
 		FixJpgOrientation: fixJpgOrientation,
 	}
 	vs.SetMasterNode(masterNode)
-	vs.store = storage.NewStore(port, ip, publicUrl, folders, maxCounts, vs.UseLevelDb)
+	vs.store = storage.NewStore(port, ip, publicUrl, folders, maxCounts, vs.needleMapKind)

 	vs.guard = security.NewGuard(whiteList, "")

@@ -77,7 +77,7 @@ func NewVolumeServer(adminMux, publicMux *http.ServeMux, ip string,
 					glog.V(0).Infoln("Volume Server Connected with master at", master)
 				}
 			} else {
-				glog.V(0).Infof("Volume Server Failed to talk with master %s: %v", vs, err)
+				glog.V(0).Infof("Volume Server Failed to talk with master %+v: %v", vs, err)
 				if connected {
 					connected = false
 				}
--- a/go/weed/weed_server/volume_server_handlers_admin.go
+++ b/go/weed/weed_server/volume_server_handlers_admin.go
@@ -17,7 +17,7 @@ func (vs *VolumeServer) statusHandler(w http.ResponseWriter, r *http.Request) {
 }

 func (vs *VolumeServer) assignVolumeHandler(w http.ResponseWriter, r *http.Request) {
-	err := vs.store.AddVolume(r.FormValue("volume"), r.FormValue("collection"), vs.UseLevelDb, r.FormValue("replication"), r.FormValue("ttl"))
+	err := vs.store.AddVolume(r.FormValue("volume"), r.FormValue("collection"), vs.needleMapKind, r.FormValue("replication"), r.FormValue("ttl"))
 	if err == nil {
 		writeJsonQuiet(w, r, http.StatusAccepted, map[string]string{"error": ""})
 	} else {