Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Improve efficiency of the scan for dirty nodes #941

Open
wants to merge 7 commits into
base: main
Choose a base branch
from
+228 −72
Open

Improve efficiency of the scan for dirty nodes #941

Show file tree
Hide file tree
Changes from 6 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
e1f2626
Add benchmark for dirty-node scan
HerbertJordan Jun 28, 2024
2161e40
Add reuse pool for dirty node candidate lists
HerbertJordan Jun 28, 2024
47b3c75
Use atomic flag to speed up dirty node check
HerbertJordan Jun 23, 2024
b470cb1
Refined implementation of atomic dirty flag
HerbertJordan Jun 28, 2024
1cf2b59
Fix node reset race condition
HerbertJordan Jun 29, 2024
37a8a05
Fix race condition in frozen node copy operation
HerbertJordan Jun 29, 2024
97f51f3
Merge branch 'main' into herbert/tune_dirty_node_scans
HerbertJordan Jul 3, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 9 additions & 8 deletions go/database/mpt/forest.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -523,10 +523,7 @@ func (s *Forest) Flush() error {
// Get snapshot of set of dirty Node IDs.
ids := make([]NodeId, 0, 1<<16)
s.nodeCache.ForEach(func(id NodeId, node *shared.Shared[Node]) {
handle := node.GetViewHandle()
dirty := handle.Get().IsDirty()
handle.Release()
if dirty {
if node.GetUnprotected().IsDirty() {
ids = append(ids, id)
}
})
Expand All @@ -553,6 +550,10 @@ func (s *Forest) flushDirtyIds(ids []NodeId) error {
if present {
handle := node.GetWriteHandle()
node := handle.Get()
if !node.IsDirty() {
handle.Release()
continue
}
err := s.flushNode(id, node)
if err == nil {
node.MarkClean()
Expand Down Expand Up @@ -910,7 +911,7 @@ func (s *Forest) createAccount() (NodeReference, shared.WriteHandle[Node], error
instance, present := s.addToCache(&ref, shared.MakeShared[Node](node))
if present {
write := instance.GetWriteHandle()
*write.Get().(*AccountNode) = *node
write.Get().Reset()
write.Release()
}
return ref, instance.GetWriteHandle(), err
Expand All @@ -926,7 +927,7 @@ func (s *Forest) createBranch() (NodeReference, shared.WriteHandle[Node], error)
instance, present := s.addToCache(&ref, shared.MakeShared[Node](node))
if present {
write := instance.GetWriteHandle()
*write.Get().(*BranchNode) = *node
write.Get().Reset()
write.Release()
}
return ref, instance.GetWriteHandle(), err
Expand All @@ -942,7 +943,7 @@ func (s *Forest) createExtension() (NodeReference, shared.WriteHandle[Node], err
instance, present := s.addToCache(&ref, shared.MakeShared[Node](node))
if present {
write := instance.GetWriteHandle()
*write.Get().(*ExtensionNode) = *node
write.Get().Reset()
write.Release()
}
return ref, instance.GetWriteHandle(), err
Expand All @@ -958,7 +959,7 @@ func (s *Forest) createValue() (NodeReference, shared.WriteHandle[Node], error)
instance, present := s.addToCache(&ref, shared.MakeShared[Node](node))
if present {
write := instance.GetWriteHandle()
*write.Get().(*ValueNode) = *node
write.Get().Reset()
write.Release()
}
return ref, instance.GetWriteHandle(), err
Expand Down
24 changes: 14 additions & 10 deletions go/database/mpt/node_flusher.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@ package mpt
import (
"errors"
"slices"
"sync"
"time"

"github.com/Fantom-foundation/Carmen/go/database/mpt/shared"
Expand Down Expand Up @@ -72,20 +73,23 @@ func (f *nodeFlusher) Stop() error {
return errors.Join(f.errs...)
}

var dirtyIdListPool = sync.Pool{
New: func() interface{} {
list := make([]NodeId, 0, 1_000_000)
return &list
},
}

func tryFlushDirtyNodes(cache NodeCache, sink NodeSink) error {
// Collect a list of dirty nodes to be flushed.
dirtyIds := make([]NodeId, 0, 1_000_000)
dirtyIdsPtr := dirtyIdListPool.Get().(*[]NodeId)
defer dirtyIdListPool.Put(dirtyIdsPtr)
dirtyIds := *(dirtyIdsPtr)
dirtyIds = dirtyIds[:0]
cache.ForEach(func(id NodeId, node *shared.Shared[Node]) {
handle, success := node.TryGetViewHandle()
if !success {
return
}
dirty := handle.Get().IsDirty()
handle.Release()
if !dirty {
return
if node.GetUnprotected().IsDirty() {
dirtyIds = append(dirtyIds, id)
}
dirtyIds = append(dirtyIds, id)
})

// The IDs are sorted to increase the chance of sequential
Expand Down
47 changes: 34 additions & 13 deletions go/database/mpt/node_flusher_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -82,7 +82,7 @@ func TestNodeFlusher_ErrorsAreCollected(t *testing.T) {
sink := NewMockNodeSink(ctrl)

id := ValueId(1)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}})

done := make(chan struct{})
counter := 0
Expand Down Expand Up @@ -115,10 +115,10 @@ func TestNodeFlusher_FlushesOnlyDirtyNodes(t *testing.T) {
sink := NewMockNodeSink(ctrl)

nodes := map[NodeId]*shared.Shared[Node]{
ValueId(1): shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: true}}),
ValueId(2): shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}}),
ValueId(3): shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}}),
ValueId(4): shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: true}}),
ValueId(1): shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusClean}}),
ValueId(2): shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}}),
ValueId(3): shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}}),
ValueId(4): shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusClean}}),
}

// All nodes are checked.
Expand Down Expand Up @@ -151,7 +151,7 @@ func TestNodeFlusher_FlushedNodesAreMarkedClean(t *testing.T) {
sink := NewMockNodeSink(ctrl)

id := ValueId(1)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}})
cache.EXPECT().ForEach(gomock.Any()).Do(func(f func(NodeId, *shared.Shared[Node])) {
f(id, node)
})
Expand All @@ -177,11 +177,13 @@ func TestNodeFlusher_NodesInUseAreIgnored(t *testing.T) {
sink := NewMockNodeSink(ctrl)

id := ValueId(1)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}})
cache.EXPECT().ForEach(gomock.Any()).Do(func(f func(NodeId, *shared.Shared[Node])) {
f(id, node)
})

cache.EXPECT().Get(RefTo(id)).Return(node, true)

// There shall be no write events (which is the default, but spelled out explicitly here).
sink.EXPECT().Write(gomock.Any(), gomock.Any()).Times(0)

Expand All @@ -206,7 +208,7 @@ func TestNodeFlusher_NodesThatAreAccessedAfterBeingIdentifiedAsDirtyAreIgnored(t
sink := NewMockNodeSink(ctrl)

id := ValueId(1)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}})
handle := shared.WriteHandle[Node]{}
cache.EXPECT().ForEach(gomock.Any()).Do(func(f func(NodeId, *shared.Shared[Node])) {
f(id, node)
Expand All @@ -232,7 +234,7 @@ func TestNodeFlusher_EvictedNodesAreIgnored(t *testing.T) {
sink := NewMockNodeSink(ctrl)

id := ValueId(1)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}})
cache.EXPECT().ForEach(gomock.Any()).Do(func(f func(NodeId, *shared.Shared[Node])) {
f(id, node)
})
Expand All @@ -256,7 +258,7 @@ func TestNodeFlusher_NodesThatGetMarkedCleanByThirdPartyAreIgnored(t *testing.T)
sink := NewMockNodeSink(ctrl)

id := ValueId(1)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}})
cache.EXPECT().ForEach(gomock.Any()).Do(func(f func(NodeId, *shared.Shared[Node])) {
f(id, node)
handle := node.GetWriteHandle()
Expand All @@ -282,7 +284,7 @@ func TestNodeFlusher_NodesWithDirtyHashesAreIgnored(t *testing.T) {
sink := NewMockNodeSink(ctrl)

id := ValueId(1)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false, hashStatus: hashStatusDirty}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty, hashStatus: hashStatusDirty}})
cache.EXPECT().ForEach(gomock.Any()).Do(func(f func(NodeId, *shared.Shared[Node])) {
f(id, node)
})
Expand All @@ -305,7 +307,7 @@ func TestNodeFlusher_FlushErrorsArePropagated(t *testing.T) {
sink := NewMockNodeSink(ctrl)

id := ValueId(1)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}})
cache.EXPECT().ForEach(gomock.Any()).Do(func(f func(NodeId, *shared.Shared[Node])) {
f(id, node)
})
Expand Down Expand Up @@ -333,7 +335,7 @@ func TestNodeFlusher_FlushErrorsAreAggregated(t *testing.T) {

id1 := ValueId(1)
id2 := ValueId(2)
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: false}})
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusDirty}})
cache.EXPECT().ForEach(gomock.Any()).Do(func(f func(NodeId, *shared.Shared[Node])) {
f(id1, node)
f(id2, node)
Expand Down Expand Up @@ -362,3 +364,22 @@ func checkThatNodeIsNotLocked(t *testing.T, node *shared.Shared[Node]) {
}
handle.Release()
}

func BenchmarkNodeFlusher_CollectDirtyNodes(b *testing.B) {
sizes := []int{1_000, 10_000, 100_000, 1_000_000, 10_000_000}
for _, size := range sizes {
b.Run(fmt.Sprintf("size=%d", size), func(b *testing.B) {
// Fill a cache with clean nodes.
cache := NewNodeCache(size)
for i := 0; i < size; i++ {
node := shared.MakeShared[Node](&ValueNode{nodeBase: nodeBase{clean: cleanStatusClean}})
ref := NewNodeReference(ValueId(uint64(i)))
cache.GetOrSet(&ref, node)
}
b.ResetTimer()
for i := 0; i < b.N; i++ {
tryFlushDirtyNodes(cache, nil)
}
})
}
}
Loading