volume layout for each replication level

weed-fs/note/startup_process.txt

@@ -1,7 +0,0 @@
-1. clients report its own server info, volumes info, 
-2. master collect all volumes, separated into readable volumes, writable volumes, volume2machine mapping
-  machines is an array of machine info
-  writable volumes is an array of vids
-  vid2machineId maps volume id to machineId, which is the index of machines array
-
-

weed-fs/src/pkg/storage/volume_info.go

@@ -1,10 +1,50 @@
 package storage
 
-import (
+import ()
-
-)
 
 type VolumeInfo struct {
-  Id   VolumeId
+	Id              VolumeId
-  Size int64
+	Size            int64
+	ReplicationType ReplicationType
+}
+type ReplicationType int
+
+const (
+  Copy00 = ReplicationType(00) // single copy
+  Copy01 = ReplicationType(01) // 2 copies, each on different racks, same data center
+  Copy10 = ReplicationType(10) // 2 copies, each on different data center
+  Copy11 = ReplicationType(11) // 3 copies, 2 on different racks and local data center, 1 on different data center
+  Copy20 = ReplicationType(20) // 3 copies, each on dffereint data center
+  LengthRelicationType = 5
+)
+
+func GetReplicationLevelIndex(v *VolumeInfo) int {
+  switch v.ReplicationType {
+  case Copy00:
+    return 0
+  case Copy01:
+    return 1
+  case Copy10:
+    return 2
+  case Copy11:
+    return 3
+  case Copy20:
+    return 4
+  }
+  return -1
+}
+func GetCopyCount(v *VolumeInfo) int {
+  switch v.ReplicationType {
+  case Copy00:
+    return 1
+  case Copy01:
+    return 2
+  case Copy10:
+    return 2
+  case Copy11:
+    return 3
+  case Copy20:
+    return 3
+  }
+  return 0
 }

weed-fs/src/pkg/topology/data_node.go

@@ -8,6 +8,10 @@ import (
 type DataNode struct {
 	NodeImpl
 	volumes   map[storage.VolumeId]*storage.VolumeInfo
+	ip        string
+	port      int
+	publicUrl string
+	lastSeen  int64 // unix time in seconds
 }
 
 func NewDataNode(id string) *DataNode {
@@ -17,12 +21,21 @@ func NewDataNode(id string) *DataNode {
 	s.volumes = make(map[storage.VolumeId]*storage.VolumeInfo)
 	return s
 }
-func (s *DataNode) CreateOneVolume(r int, vid storage.VolumeId) storage.VolumeId {
+func (dn *DataNode) CreateOneVolume(r int, vid storage.VolumeId) storage.VolumeId {
-	s.AddVolume(&storage.VolumeInfo{Id: vid, Size: 32 * 1024 * 1024 * 1024})
+	dn.AddVolume(&storage.VolumeInfo{Id: vid, Size: 32 * 1024 * 1024 * 1024})
 	return vid
 }
-func (s *DataNode) AddVolume(v *storage.VolumeInfo) {
+func (dn *DataNode) AddVolume(v *storage.VolumeInfo) {
-	s.volumes[v.Id] = v
+	dn.volumes[v.Id] = v
-	s.UpAdjustActiveVolumeCountDelta(1)
+	dn.UpAdjustActiveVolumeCountDelta(1)
-	s.UpAdjustMaxVolumeId(v.Id)
+	dn.UpAdjustMaxVolumeId(v.Id)
+	dn.GetTopology().RegisterVolume(v,dn)
+}
+func (dn *DataNode) GetTopology() *Topology {
+  p := dn.parent
+  for p.Parent()!=nil{
+    p = p.Parent()
+  }
+  t := p.(*Topology)
+  return t
 }

weed-fs/src/pkg/topology/node.go

@@ -20,10 +20,11 @@ type Node interface {
 	setParent(Node)
 	LinkChildNode(node Node)
 	UnlinkChildNode(nodeId NodeId)
+	CollectWritableVolumes(freshThreshHold int64, volumeSizeLimit uint64) []storage.VolumeId
 
 	IsDataNode() bool
 	Children() map[NodeId]Node
-  Parent() Node
+	Parent() Node
 }
 type NodeImpl struct {
 	id                NodeId
@@ -65,7 +66,7 @@ func (n *NodeImpl) Children() map[NodeId]Node {
 	return n.children
 }
 func (n *NodeImpl) Parent() Node {
-  return n.parent
+	return n.parent
 }
 func (n *NodeImpl) ReserveOneVolume(r int, vid storage.VolumeId) (bool, *DataNode) {
 	ret := false
@@ -146,3 +147,26 @@ func (n *NodeImpl) UnlinkChildNode(nodeId NodeId) {
 		fmt.Println(n, "removes", node, "volumeCount =", n.activeVolumeCount)
 	}
 }
+
+func (n *NodeImpl) CollectWritableVolumes(freshThreshHold int64, volumeSizeLimit uint64) []storage.VolumeId {
+	var ret []storage.VolumeId
+	if n.IsRack() {
+		for _, c := range n.Children() {
+			dn := c.(*DataNode) //can not cast n to DataNode
+			if dn.lastSeen > freshThreshHold {
+			  continue
+			}
+			for _, v := range dn.volumes {
+				if uint64(v.Size) < volumeSizeLimit {
+					ret = append(ret, v.Id)
+				}
+			}
+		}
+	} else {
+		for _, c := range n.Children() {
+			ret = append(ret, c.CollectWritableVolumes(freshThreshHold, volumeSizeLimit)...)
+		}
+	}
+
+	return ret
+}

weed-fs/src/pkg/topology/topology.go

@@ -3,37 +3,59 @@ package topology
 import (
 	_ "fmt"
 	"math/rand"
+	"pkg/sequence"
 	"pkg/storage"
 )
 
 type Topology struct {
 	NodeImpl
+
+	//transient vid~servers mapping for each replication type
+	replicaType2VolumeLayout []*VolumeLayout
+
+	pulse int64
+
+	volumeSizeLimit uint64
+
+	sequence sequence.Sequencer
 }
 
-func NewTopology(id string) *Topology {
+func NewTopology(id string, dirname string, filename string, volumeSizeLimit uint64, pulse int) *Topology {
 	t := &Topology{}
 	t.id = NodeId(id)
 	t.nodeType = "Topology"
 	t.children = make(map[NodeId]Node)
+	t.replicaType2VolumeLayout = make([]*VolumeLayout, storage.LengthRelicationType)
+	t.pulse = int64(pulse)
+	t.volumeSizeLimit = volumeSizeLimit
+	t.sequence = sequence.NewSequencer(dirname, filename)
 	return t
 }
 func (t *Topology) RandomlyReserveOneVolume() (bool, *DataNode, storage.VolumeId) {
 	vid := t.NextVolumeId()
 	ret, node := t.ReserveOneVolume(rand.Intn(t.FreeSpace()), vid)
-  return ret, node, vid
+	return ret, node, vid
 }
 
 func (t *Topology) RandomlyReserveOneVolumeExcept(except []Node) (bool, *DataNode, storage.VolumeId) {
-  freeSpace := t.FreeSpace()
+	freeSpace := t.FreeSpace()
-  for _, node := range except {
+	for _, node := range except {
-    freeSpace -= node.FreeSpace()
+		freeSpace -= node.FreeSpace()
-  }
+	}
-  vid := t.NextVolumeId()
+	vid := t.NextVolumeId()
-  ret, node := t.ReserveOneVolume(rand.Intn(freeSpace), vid)
+	ret, node := t.ReserveOneVolume(rand.Intn(freeSpace), vid)
-  return ret, node, vid
+	return ret, node, vid
 }
 
 func (t *Topology) NextVolumeId() storage.VolumeId {
 	vid := t.GetMaxVolumeId()
 	return vid.Next()
 }
+
+func (t *Topology) RegisterVolume(v *storage.VolumeInfo, dn *DataNode) {
+	replicationTypeIndex := storage.GetReplicationLevelIndex(v)
+	if t.replicaType2VolumeLayout[replicationTypeIndex] == nil {
+		t.replicaType2VolumeLayout[replicationTypeIndex] = NewVolumeLayout(t.volumeSizeLimit, t.pulse)
+	}
+	t.replicaType2VolumeLayout[replicationTypeIndex].RegisterVolume(v, dn)
+}

weed-fs/src/pkg/topology/volume_layout.go

@@ -0,0 +1,48 @@
+package topology
+
+import (
+  "errors"
+	"fmt"
+	"math/rand"
+	"pkg/storage"
+)
+
+type VolumeLayout struct {
+	vid2location    map[storage.VolumeId]*DataNodeLocationList
+	writables       []storage.VolumeId // transient array of writable volume id
+	pulse           int64
+	volumeSizeLimit uint64
+}
+
+func NewVolumeLayout(volumeSizeLimit uint64, pulse int64) *VolumeLayout {
+	return &VolumeLayout{
+		vid2location:    make(map[storage.VolumeId]*DataNodeLocationList),
+		writables:       *new([]storage.VolumeId),
+		pulse:           pulse,
+		volumeSizeLimit: volumeSizeLimit,
+	}
+}
+
+func (vl *VolumeLayout) RegisterVolume(v *storage.VolumeInfo, dn *DataNode) {
+	if _, ok := vl.vid2location[v.Id]; !ok {
+		vl.vid2location[v.Id] = NewDataNodeLocationList()
+	}
+	vl.vid2location[v.Id].Add(dn)
+	if len(vl.vid2location[v.Id].list) >= storage.GetCopyCount(v) {
+	  vl.writables = append(vl.writables,v.Id)
+	}
+}
+
+func (vl *VolumeLayout) PickForWrite(count int) (int, *DataNodeLocationList, error) {
+  len_writers := len(vl.writables)
+  if len_writers <= 0 {
+    fmt.Println("No more writable volumes!")
+    return 0, nil, errors.New("No more writable volumes!")
+  }
+  vid := vl.writables[rand.Intn(len_writers)]
+  locationList := vl.vid2location[vid]
+  if locationList != nil {
+    return count, locationList, nil
+  }
+  return 0, nil, errors.New("Strangely vid " + vid.String() + " is on no machine!")
+}

weed-fs/src/pkg/topology/volume_location.go

@@ -0,0 +1,21 @@
+package topology
+
+import (
+)
+
+type DataNodeLocationList struct {
+  list []*DataNode
+}
+
+func NewDataNodeLocationList() *DataNodeLocationList {
+  return &DataNodeLocationList{}
+}
+
+func (dnll *DataNodeLocationList) Add(loc *DataNode){
+  for _, dnl := range dnll.list {
+    if loc.ip == dnl.ip && loc.port == dnl.port {
+      break
+    }
+  }
+  dnll.list = append(dnll.list, loc)
+}