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

db: add IteratorStats, for counting the work done in Iterator #1129

Merged
merged 1 commit into from
May 5, 2021
Merged

db: add IteratorStats, for counting the work done in Iterator #1129

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
84 changes: 83 additions & 1 deletion iterator.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@ import (
"github.com/cockroachdb/pebble/internal/fastrand"
"github.com/cockroachdb/pebble/internal/invariants"
"github.com/cockroachdb/pebble/internal/manifest"
"github.com/cockroachdb/redact"
)

// iterPos describes the state of the internal iterator, in terms of whether it is
Expand Down Expand Up @@ -49,7 +50,8 @@ const readBytesPeriod uint64 = 1 << 16

var errReversePrefixIteration = errors.New("pebble: unsupported reverse prefix iteration")

// IteratorMetrics holds per-iterator metrics.
// IteratorMetrics holds per-iterator metrics. These do not change over the
// lifetime of the iterator.
type IteratorMetrics struct {
// The read amplification experienced by this iterator. This is the sum of
// the memtables, the L0 sublevels and the non-empty Ln levels. Higher read
Expand All @@ -58,6 +60,32 @@ type IteratorMetrics struct {
ReadAmp int
}

// IteratorStatsKind describes the two kind of iterator stats.
type IteratorStatsKind int8

const (
// InterfaceCall represents calls to Iterator.
InterfaceCall IteratorStatsKind = iota
// InternalIterCall represents calls by Iterator to its internalIterator.
InternalIterCall
// NumStatsKind is the number of kinds, and is used for array sizing.
NumStatsKind
)

// IteratorStats contains iteration stats.
type IteratorStats struct {
// ForwardSeekCount includes SeekGE, SeekPrefixGE, First.
ForwardSeekCount [NumStatsKind]int
// ReverseSeek includes SeekLT, Last.
ReverseSeekCount [NumStatsKind]int
// ForwardStepCount includes Next.
ForwardStepCount [NumStatsKind]int
// ReverseStepCount includes Prev.
ReverseStepCount [NumStatsKind]int
}

var _ redact.SafeFormatter = &IteratorStats{}

// Iterator iterates over a DB's key/value pairs in key order.
//
// An iterator must be closed after use, but it is not necessary to read an
Expand Down Expand Up @@ -101,6 +129,7 @@ type Iterator struct {
getIterAlloc *getIterAlloc
prefixOrFullSeekKey []byte
readSampling readSampling
stats IteratorStats

// Following fields are only used in Clone.
// Non-nil if this Iterator includes a Batch.
Expand Down Expand Up @@ -257,6 +286,7 @@ func (i *Iterator) nextUserKey() {
}
for {
i.iterKey, i.iterValue = i.iter.Next()
i.stats.ForwardStepCount[InternalIterCall]++
if done || i.iterKey == nil {
break
}
Expand Down Expand Up @@ -412,6 +442,7 @@ func (i *Iterator) findPrevEntry(limit []byte) {
i.iterValidityState = IterExhausted
valueMerger = nil
i.iterKey, i.iterValue = i.iter.Prev()
i.stats.ReverseStepCount[InternalIterCall]++
// Compare with the limit. We could optimize by only checking when
// we step to the previous user key, but detecting that requires a
// comparison too. Note that this position may already passed a
Expand Down Expand Up @@ -439,6 +470,7 @@ func (i *Iterator) findPrevEntry(limit []byte) {
i.value = i.valueBuf
i.iterValidityState = IterValid
i.iterKey, i.iterValue = i.iter.Prev()
i.stats.ReverseStepCount[InternalIterCall]++
valueMerger = nil
continue

Expand Down Expand Up @@ -468,6 +500,7 @@ func (i *Iterator) findPrevEntry(limit []byte) {
}
}
i.iterKey, i.iterValue = i.iter.Prev()
i.stats.ReverseStepCount[InternalIterCall]++
continue

default:
Expand Down Expand Up @@ -501,6 +534,7 @@ func (i *Iterator) prevUserKey() {
}
for {
i.iterKey, i.iterValue = i.iter.Prev()
i.stats.ReverseStepCount[InternalIterCall]++
if i.iterKey == nil {
break
}
Expand All @@ -519,6 +553,7 @@ func (i *Iterator) mergeNext(key InternalKey, valueMerger ValueMerger) {
// Loop looking for older values for this key and merging them.
for {
i.iterKey, i.iterValue = i.iter.Next()
i.stats.ForwardStepCount[InternalIterCall]++
if i.iterKey == nil {
i.pos = iterPosNext
return
Expand Down Expand Up @@ -572,6 +607,7 @@ func (i *Iterator) SeekGEWithLimit(key []byte, limit []byte) IterValidityState {
i.lastPositioningOp = unknownLastPositionOp
i.err = nil // clear cached iteration error
i.hasPrefix = false
i.stats.ForwardSeekCount[InterfaceCall]++
if lowerBound := i.opts.GetLowerBound(); lowerBound != nil && i.cmp(key, lowerBound) < 0 {
key = lowerBound
} else if upperBound := i.opts.GetUpperBound(); upperBound != nil && i.cmp(key, upperBound) > 0 {
Expand Down Expand Up @@ -606,6 +642,7 @@ func (i *Iterator) SeekGEWithLimit(key []byte, limit []byte) IterValidityState {
}
if seekInternalIter {
i.iterKey, i.iterValue = i.iter.SeekGE(key)
i.stats.ForwardSeekCount[InternalIterCall]++
}
i.findNextEntry(limit)
i.maybeSampleRead()
Expand Down Expand Up @@ -667,6 +704,7 @@ func (i *Iterator) SeekPrefixGE(key []byte) bool {
// iterator position.
i.lastPositioningOp = unknownLastPositionOp
i.err = nil // clear cached iteration error
i.stats.ForwardSeekCount[InterfaceCall]++

if i.split == nil {
panic("pebble: split must be provided for SeekPrefixGE")
Expand Down Expand Up @@ -728,6 +766,7 @@ func (i *Iterator) SeekPrefixGE(key []byte) bool {
}

i.iterKey, i.iterValue = i.iter.SeekPrefixGE(i.prefixOrFullSeekKey, key, trySeekUsingNext)
i.stats.ForwardSeekCount[InternalIterCall]++
i.findNextEntry(nil)
i.maybeSampleRead()
if i.Error() == nil {
Expand Down Expand Up @@ -761,6 +800,7 @@ func (i *Iterator) SeekLTWithLimit(key []byte, limit []byte) IterValidityState {
i.lastPositioningOp = unknownLastPositionOp
i.err = nil // clear cached iteration error
i.hasPrefix = false
i.stats.ReverseSeekCount[InterfaceCall]++
if upperBound := i.opts.GetUpperBound(); upperBound != nil && i.cmp(key, upperBound) > 0 {
key = upperBound
} else if lowerBound := i.opts.GetLowerBound(); lowerBound != nil && i.cmp(key, lowerBound) < 0 {
Expand Down Expand Up @@ -797,6 +837,7 @@ func (i *Iterator) SeekLTWithLimit(key []byte, limit []byte) IterValidityState {
}
if seekInternalIter {
i.iterKey, i.iterValue = i.iter.SeekLT(key)
i.stats.ReverseSeekCount[InternalIterCall]++
}
i.findPrevEntry(limit)
i.maybeSampleRead()
Expand All @@ -814,10 +855,13 @@ func (i *Iterator) First() bool {
i.err = nil // clear cached iteration error
i.hasPrefix = false
i.lastPositioningOp = unknownLastPositionOp
i.stats.ForwardSeekCount[InterfaceCall]++
if lowerBound := i.opts.GetLowerBound(); lowerBound != nil {
i.iterKey, i.iterValue = i.iter.SeekGE(lowerBound)
i.stats.ForwardSeekCount[InternalIterCall]++
} else {
i.iterKey, i.iterValue = i.iter.First()
i.stats.ForwardSeekCount[InternalIterCall]++
}
i.findNextEntry(nil)
i.maybeSampleRead()
Expand All @@ -830,10 +874,13 @@ func (i *Iterator) Last() bool {
i.err = nil // clear cached iteration error
i.hasPrefix = false
i.lastPositioningOp = unknownLastPositionOp
i.stats.ReverseSeekCount[InterfaceCall]++
if upperBound := i.opts.GetUpperBound(); upperBound != nil {
i.iterKey, i.iterValue = i.iter.SeekLT(upperBound)
i.stats.ReverseSeekCount[InternalIterCall]++
} else {
i.iterKey, i.iterValue = i.iter.Last()
i.stats.ReverseSeekCount[InternalIterCall]++
}
i.findPrevEntry(nil)
i.maybeSampleRead()
Expand All @@ -848,6 +895,7 @@ func (i *Iterator) Next() bool {

// NextWithLimit ...
func (i *Iterator) NextWithLimit(limit []byte) IterValidityState {
i.stats.ForwardStepCount[InterfaceCall]++
if limit != nil && i.hasPrefix {
i.err = errors.New("cannot use limit with prefix iteration")
i.iterValidityState = IterExhausted
Expand Down Expand Up @@ -875,8 +923,10 @@ func (i *Iterator) NextWithLimit(limit []byte) IterValidityState {
// the first key.
if lowerBound := i.opts.GetLowerBound(); lowerBound != nil {
i.iterKey, i.iterValue = i.iter.SeekGE(lowerBound)
i.stats.ForwardSeekCount[InternalIterCall]++
} else {
i.iterKey, i.iterValue = i.iter.First()
i.stats.ForwardSeekCount[InternalIterCall]++
}
case iterPosCurReversePaused:
// Switching directions.
Expand All @@ -899,8 +949,10 @@ func (i *Iterator) NextWithLimit(limit []byte) IterValidityState {
// the first key.
if lowerBound := i.opts.GetLowerBound(); lowerBound != nil {
i.iterKey, i.iterValue = i.iter.SeekGE(lowerBound)
i.stats.ForwardSeekCount[InternalIterCall]++
} else {
i.iterKey, i.iterValue = i.iter.First()
i.stats.ForwardSeekCount[InternalIterCall]++
}
} else {
i.nextUserKey()
Expand All @@ -922,6 +974,7 @@ func (i *Iterator) Prev() bool {

// PrevWithLimit ...
func (i *Iterator) PrevWithLimit(limit []byte) IterValidityState {
i.stats.ReverseStepCount[InterfaceCall]++
if i.err != nil {
return i.iterValidityState
}
Expand Down Expand Up @@ -958,8 +1011,10 @@ func (i *Iterator) PrevWithLimit(limit []byte) IterValidityState {
// the last key.
if upperBound := i.opts.GetUpperBound(); upperBound != nil {
i.iterKey, i.iterValue = i.iter.SeekLT(upperBound)
i.stats.ReverseSeekCount[InternalIterCall]++
} else {
i.iterKey, i.iterValue = i.iter.Last()
i.stats.ReverseSeekCount[InternalIterCall]++
}
} else {
i.prevUserKey()
Expand Down Expand Up @@ -1115,6 +1170,16 @@ func (i *Iterator) Metrics() IteratorMetrics {
return m
}

// ResetStats resets the stats to 0.
func (i *Iterator) ResetStats() {
i.stats = IteratorStats{}
}

// Stats returns the current stats.
func (i *Iterator) Stats() IteratorStats {
return i.stats
}

// Clone creates a new Iterator over the same underlying data, i.e., over the
// same {batch, memtables, sstables}). It starts with the same IterOptions but
// is not positioned. Note that IterOptions is not deep-copied, so the
Expand Down Expand Up @@ -1161,3 +1226,20 @@ func (i *Iterator) Clone() (*Iterator, error) {
}
return finishInitializingIter(buf), nil
}

func (stats *IteratorStats) String() string {
return redact.StringWithoutMarkers(stats)
}

// SafeFormat implements the redact.SafeFormatter interface.
func (stats *IteratorStats) SafeFormat(s redact.SafePrinter, verb rune) {
for i := range stats.ForwardStepCount {
switch IteratorStatsKind(i) {
case InterfaceCall: s.SafeString("(interface (dir, seek, step): ")
case InternalIterCall: s.SafeString(", (internal (dir, seek, step): ")
}
s.Printf("(fwd, %d, %d), (rev, %d, %d))",
redact.Safe(stats.ForwardSeekCount[i]), redact.Safe(stats.ForwardStepCount[i]),
redact.Safe(stats.ReverseSeekCount[i]), redact.Safe(stats.ReverseStepCount[i]))
}
}
4 changes: 3 additions & 1 deletion iterator_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -473,7 +473,9 @@ func TestIterator(t *testing.T) {
}

iter := newIter(uint64(seqNum), opts)
return runIterCmd(d, iter, true)
iterOutput := runIterCmd(d, iter, true)
stats := iter.Stats()
return fmt.Sprintf("%sstats: %s\n", iterOutput, stats.String())

default:
return fmt.Sprintf("unknown command: %s", d.Cmd)
Expand Down
Loading