ingest,compaction: use excise when flushing flushableIngest

This patch amends the ingest pipeline to allow for certain cases to use
`flushaleIngest` in order to avoid waiting on memtable flushes when
there is overlap. The current use for this is range snapshots in
Cockroach where we include `RANGEDEL`. When we use this path, we also
pass the `exciseSpan` down down into the flushable to then excise at
flush time to still benefit from the use of excise in the ingest path.

Fixes cockroachdb#3335.

compaction.go

@@ -1448,17 +1448,70 @@ func (d *DB) runIngestFlush(c *compaction) (*manifest.VersionEdit, error) {
 	var err error
 	var fileToSplit *fileMetadata
 	var ingestSplitFiles []ingestSplitFile
-	for _, file := range c.flushing[0].flushable.(*ingestedFlushable).files {
-		suggestSplit := d.opts.Experimental.IngestSplit != nil && d.opts.Experimental.IngestSplit() &&
-			d.FormatMajorVersion() >= FormatVirtualSSTables
-		level, fileToSplit, err = ingestTargetLevel(
-			d.newIters, d.tableNewRangeKeyIter, iterOpts, d.opts.Comparer,
-			c.version, baseLevel, d.mu.compact.inProgress, file.FileMetadata,
-			suggestSplit,
-		)
+	ingestFlushable := c.flushing[0].flushable.(*ingestedFlushable)
+
+	updateLevelMetricsOnExcise := func(m *fileMetadata, level int, added []newFileEntry) {
+		levelMetrics := c.metrics[level]
+		if levelMetrics == nil {
+			levelMetrics = &LevelMetrics{}
+			c.metrics[level] = levelMetrics
+		}
+		levelMetrics.NumFiles--
+		levelMetrics.Size -= int64(m.Size)
+		for i := range added {
+			levelMetrics.NumFiles++
+			levelMetrics.Size += int64(added[i].Meta.Size)
+		}
+	}
+
+	suggestSplit := d.opts.Experimental.IngestSplit != nil && d.opts.Experimental.IngestSplit() &&
+		d.FormatMajorVersion() >= FormatVirtualSSTables
+
+	replacedFiles := make(map[base.FileNum][]newFileEntry)
+	for _, file := range ingestFlushable.files {
+		// This file fits perfectly within the excise span, so we can slot it at L6.
+		if ingestFlushable.exciseSpan.Valid() &&
+			ingestFlushable.exciseSpan.Contains(d.cmp, file.FileMetadata.Smallest) &&
+			ingestFlushable.exciseSpan.Contains(d.cmp, file.FileMetadata.Largest) {
+			level = 6
+		} else {
+			level, fileToSplit, err = ingestTargetLevel(
+				d.newIters, d.tableNewRangeKeyIter, iterOpts, d.opts.Comparer,
+				c.version, baseLevel, d.mu.compact.inProgress, file.FileMetadata,
+				suggestSplit,
+			)
+		}
+
+		if ingestFlushable.exciseSpan.Valid() {
+			ve.DeletedFiles = map[manifest.DeletedFileEntry]*manifest.FileMetadata{}
+			// Iterate through all levels and find files that intersect with exciseSpan.
+			for level = range c.version.Levels {
+				overlaps := c.version.Overlaps(level, base.UserKeyBoundsEndExclusive(ingestFlushable.exciseSpan.Start, ingestFlushable.exciseSpan.End))
+				iter := overlaps.Iter()
+
+				for m := iter.First(); m != nil; m = iter.Next() {
+					newFiles, err := d.excise(ingestFlushable.exciseSpan, m, ve, level)
+					if err != nil {
+						return nil, err
+					}
+
+					if _, ok := ve.DeletedFiles[deletedFileEntry{
+						Level:   level,
+						FileNum: m.FileNum,
+					}]; !ok {
+						// We did not excise this file.
+						continue
+					}
+					replacedFiles[m.FileNum] = newFiles
+					updateLevelMetricsOnExcise(m, level, newFiles)
+				}
+			}
+		}
 		if err != nil {
 			return nil, err
 		}
+
+		// Add the current flushableIngest file to the version.
 		ve.NewFiles = append(ve.NewFiles, newFileEntry{Level: level, Meta: file.FileMetadata})
 		if fileToSplit != nil {
 			ingestSplitFiles = append(ingestSplitFiles, ingestSplitFile{
@@ -1476,23 +1529,7 @@ func (d *DB) runIngestFlush(c *compaction) (*manifest.VersionEdit, error) {
 		levelMetrics.TablesIngested++
 	}
 
-	updateLevelMetricsOnExcise := func(m *fileMetadata, level int, added []newFileEntry) {
-		levelMetrics := c.metrics[level]
-		if levelMetrics == nil {
-			levelMetrics = &LevelMetrics{}
-			c.metrics[level] = levelMetrics
-		}
-		levelMetrics.NumFiles--
-		levelMetrics.Size -= int64(m.Size)
-		for i := range added {
-			levelMetrics.NumFiles++
-			levelMetrics.Size += int64(added[i].Meta.Size)
-		}
-	}
-
 	if len(ingestSplitFiles) > 0 {
-		ve.DeletedFiles = make(map[manifest.DeletedFileEntry]*manifest.FileMetadata)
-		replacedFiles := make(map[base.FileNum][]newFileEntry)
 		if err := d.ingestSplit(ve, updateLevelMetricsOnExcise, ingestSplitFiles, replacedFiles); err != nil {
 			return nil, err
 		}

data_test.go

@@ -1258,6 +1258,7 @@ func (d *DB) waitTableStats() {
 
 func runIngestAndExciseCmd(td *datadriven.TestData, d *DB, fs vfs.FS) error {
 	var exciseSpan KeyRange
+	var sstContainsExciseTombstone bool
 	paths := make([]string, 0, len(td.CmdArgs))
 	for i, arg := range td.CmdArgs {
 		switch td.CmdArgs[i].Key {
@@ -1271,12 +1272,14 @@ func runIngestAndExciseCmd(td *datadriven.TestData, d *DB, fs vfs.FS) error {
 			}
 			exciseSpan.Start = []byte(fields[0])
 			exciseSpan.End = []byte(fields[1])
+		case "contains-excise-tombstone":
+			sstContainsExciseTombstone = true
 		default:
 			paths = append(paths, arg.String())
 		}
 	}
 
-	if _, err := d.IngestAndExcise(paths, nil /* shared */, nil /* external */, exciseSpan); err != nil {
+	if _, err := d.IngestAndExcise(paths, nil /* shared */, nil /* external */, exciseSpan, sstContainsExciseTombstone); err != nil {
 		return err
 	}
 	return nil

flushable.go

@@ -162,13 +162,17 @@ type ingestedFlushable struct {
 	slice manifest.LevelSlice
 	// hasRangeKeys is set on ingestedFlushable construction.
 	hasRangeKeys bool
+	// exciseSpan is populated if an excise operation should be performed during
+	// flush.
+	exciseSpan KeyRange
 }
 
 func newIngestedFlushable(
 	files []*fileMetadata,
 	comparer *Comparer,
 	newIters tableNewIters,
 	newRangeKeyIters keyspanimpl.TableNewSpanIter,
+	exciseSpan KeyRange,
 ) *ingestedFlushable {
 	if invariants.Enabled {
 		for i := 1; i < len(files); i++ {
@@ -196,6 +200,7 @@ func newIngestedFlushable(
 		// slice is immutable and can be set once and used many times.
 		slice:        manifest.NewLevelSliceKeySorted(comparer.Compare, files),
 		hasRangeKeys: hasRangeKeys,
+		exciseSpan:   exciseSpan,
 	}
 
 	return ret

flushable_test.go

@@ -119,9 +119,7 @@ func TestIngestedSSTFlushableAPI(t *testing.T) {
 			}
 
 			meta := loadFileMeta(paths)
-			flushable = newIngestedFlushable(
-				meta, d.opts.Comparer, d.newIters, d.tableNewRangeKeyIter,
-			)
+			flushable = newIngestedFlushable(meta, d.opts.Comparer, d.newIters, d.tableNewRangeKeyIter, KeyRange{})
 			return ""
 		case "iter":
 			iter := flushable.newIter(nil)

ingest.go

@@ -1162,7 +1162,7 @@ func (d *DB) Ingest(paths []string) error {
 	if d.opts.ReadOnly {
 		return ErrReadOnly
 	}
-	_, err := d.ingest(paths, ingestTargetLevel, nil /* shared */, KeyRange{}, nil /* external */)
+	_, err := d.ingest(paths, ingestTargetLevel, nil /* shared */, KeyRange{}, false, nil /* external */)
 	return err
 }
 
@@ -1250,7 +1250,7 @@ func (d *DB) IngestWithStats(paths []string) (IngestOperationStats, error) {
 	if d.opts.ReadOnly {
 		return IngestOperationStats{}, ErrReadOnly
 	}
-	return d.ingest(paths, ingestTargetLevel, nil /* shared */, KeyRange{}, nil /* external */)
+	return d.ingest(paths, ingestTargetLevel, nil, KeyRange{}, false, nil)
 }
 
 // IngestExternalFiles does the same as IngestWithStats, and additionally
@@ -1268,7 +1268,7 @@ func (d *DB) IngestExternalFiles(external []ExternalFile) (IngestOperationStats,
 	if d.opts.Experimental.RemoteStorage == nil {
 		return IngestOperationStats{}, errors.New("pebble: cannot ingest external files without shared storage configured")
 	}
-	return d.ingest(nil, ingestTargetLevel, nil /* shared */, KeyRange{}, external)
+	return d.ingest(nil, ingestTargetLevel, nil, KeyRange{}, false, external)
 }
 
 // IngestAndExcise does the same as IngestWithStats, and additionally accepts a
@@ -1282,7 +1282,11 @@ func (d *DB) IngestExternalFiles(external []ExternalFile) (IngestOperationStats,
 // Panics if this DB instance was not instantiated with a remote.Storage and
 // shared sstables are present.
 func (d *DB) IngestAndExcise(
-	paths []string, shared []SharedSSTMeta, external []ExternalFile, exciseSpan KeyRange,
+	paths []string,
+	shared []SharedSSTMeta,
+	external []ExternalFile,
+	exciseSpan KeyRange,
+	sstsContainExciseTombstone bool,
 ) (IngestOperationStats, error) {
 	if err := d.closed.Load(); err != nil {
 		panic(err)
@@ -1305,12 +1309,12 @@ func (d *DB) IngestAndExcise(
 			v, FormatMinForSharedObjects,
 		)
 	}
-	return d.ingest(paths, ingestTargetLevel, shared, exciseSpan, external)
+	return d.ingest(paths, ingestTargetLevel, shared, exciseSpan, sstsContainExciseTombstone, external)
 }
 
 // Both DB.mu and commitPipeline.mu must be held while this is called.
 func (d *DB) newIngestedFlushableEntry(
-	meta []*fileMetadata, seqNum uint64, logNum base.DiskFileNum,
+	meta []*fileMetadata, seqNum uint64, logNum base.DiskFileNum, exciseSpan KeyRange,
 ) (*flushableEntry, error) {
 	// Update the sequence number for all of the sstables in the
 	// metadata. Writing the metadata to the manifest when the
@@ -1326,7 +1330,7 @@ func (d *DB) newIngestedFlushableEntry(
 		}
 	}
 
-	f := newIngestedFlushable(meta, d.opts.Comparer, d.newIters, d.tableNewRangeKeyIter)
+	f := newIngestedFlushable(meta, d.opts.Comparer, d.newIters, d.tableNewRangeKeyIter, exciseSpan)
 
 	// NB: The logNum/seqNum are the WAL number which we're writing this entry
 	// to and the sequence number within the WAL which we'll write this entry
@@ -1356,7 +1360,9 @@ func (d *DB) newIngestedFlushableEntry(
 // we're holding both locks, the order in which we rotate the memtable or
 // recycle the WAL in this function is irrelevant as long as the correct log
 // numbers are assigned to the appropriate flushable.
-func (d *DB) handleIngestAsFlushable(meta []*fileMetadata, seqNum uint64) error {
+func (d *DB) handleIngestAsFlushable(
+	meta []*fileMetadata, seqNum uint64, exciseSpan KeyRange,
+) error {
 	b := d.NewBatch()
 	for _, m := range meta {
 		b.ingestSST(m.FileNum)
@@ -1382,7 +1388,7 @@ func (d *DB) handleIngestAsFlushable(meta []*fileMetadata, seqNum uint64) error
 		d.mu.Lock()
 	}
 
-	entry, err := d.newIngestedFlushableEntry(meta, seqNum, logNum)
+	entry, err := d.newIngestedFlushableEntry(meta, seqNum, logNum, exciseSpan)
 	if err != nil {
 		return err
 	}
@@ -1411,6 +1417,7 @@ func (d *DB) handleIngestAsFlushable(meta []*fileMetadata, seqNum uint64) error
 	d.mu.mem.queue = append(d.mu.mem.queue, entry)
 	d.rotateMemtable(newLogNum, nextSeqNum, currMem)
 	d.updateReadStateLocked(d.opts.DebugCheck)
+	// TODO(aaditya): is this necessary? we call this already in rotateMemtable above
 	d.maybeScheduleFlush()
 	return nil
 }
@@ -1421,6 +1428,7 @@ func (d *DB) ingest(
 	targetLevelFunc ingestTargetLevelFunc,
 	shared []SharedSSTMeta,
 	exciseSpan KeyRange,
+	sstsContainExciseTombstone bool,
 	external []ExternalFile,
 ) (IngestOperationStats, error) {
 	if len(shared) > 0 && d.opts.Experimental.RemoteStorage == nil {
@@ -1621,11 +1629,19 @@ func (d *DB) ingest(
 			mut.writerRef()
 			return
 		}
-		// The ingestion overlaps with some entry in the flushable queue.
-		if d.FormatMajorVersion() < FormatFlushableIngest ||
-			d.opts.Experimental.DisableIngestAsFlushable() ||
-			len(shared) > 0 || exciseSpan.Valid() || len(external) > 0 ||
-			(len(d.mu.mem.queue) > d.opts.MemTableStopWritesThreshold-1) {
+
+		// The ingestion overlaps with some entry in the flushable queue. If the
+		// pre-conditions are met below, we can treat this ingestion as a flushable
+		// ingest, otherwise we wait on the memtable flush before ingestion.
+		//
+		// TODO(aaditya): We should make flushableIngest compatible with remote
+		// files.
+		hasRemoteFiles := len(shared) > 0 || len(external) > 0
+		canIngestFlushable := d.FormatMajorVersion() >= FormatFlushableIngest &&
+			(len(d.mu.mem.queue) < d.opts.MemTableStopWritesThreshold) &&
+			!d.opts.Experimental.DisableIngestAsFlushable() && !hasRemoteFiles
+
+		if !canIngestFlushable || (exciseSpan.Valid() && !sstsContainExciseTombstone) {
 			// We're not able to ingest as a flushable,
 			// so we must synchronously flush.
 			//
@@ -1657,7 +1673,7 @@ func (d *DB) ingest(
 		for i := range fileMetas {
 			fileMetas[i] = loadResult.local[i].fileMetadata
 		}
-		err = d.handleIngestAsFlushable(fileMetas, seqNum)
+		err = d.handleIngestAsFlushable(fileMetas, seqNum, exciseSpan)
 	}
 
 	var ve *versionEdit

ingest_test.go

@@ -696,6 +696,9 @@ func TestExcise(t *testing.T) {
 
 		case "ingest-and-excise":
 			flushed = false
+			d.mu.Lock()
+			prevFlushableIngests := d.mu.versions.metrics.Flush.AsIngestCount
+			d.mu.Unlock()
 			if err := runIngestAndExciseCmd(td, d, mem); err != nil {
 				return err.Error()
 			}
@@ -704,8 +707,12 @@ func TestExcise(t *testing.T) {
 			for d.mu.compact.flushing {
 				d.mu.compact.cond.Wait()
 			}
+			flushableIngests := d.mu.versions.metrics.Flush.AsIngestCount
 			d.mu.Unlock()
 			if flushed {
+				if prevFlushableIngests < flushableIngests {
+					return "flushable ingest"
+				}
 				return "memtable flushed"
 			}
 			return ""
@@ -1091,7 +1098,7 @@ func testIngestSharedImpl(
 			require.NoError(t, err)
 			require.NoError(t, w.Close())
 
-			_, err = to.IngestAndExcise([]string{sstPath}, sharedSSTs, nil /* external */, KeyRange{Start: startKey, End: endKey})
+			_, err = to.IngestAndExcise([]string{sstPath}, sharedSSTs, nil /* external */, KeyRange{Start: startKey, End: endKey}, false)
 			require.NoError(t, err)
 			return fmt.Sprintf("replicated %d shared SSTs", len(sharedSSTs))
 
@@ -1320,7 +1327,7 @@ func TestSimpleIngestShared(t *testing.T) {
 		Level:            6,
 		Size:             uint64(size + 5),
 	}
-	_, err = d.IngestAndExcise([]string{}, []SharedSSTMeta{sharedSSTMeta}, nil /* external */, KeyRange{Start: []byte("d"), End: []byte("ee")})
+	_, err = d.IngestAndExcise([]string{}, []SharedSSTMeta{sharedSSTMeta}, nil /* external */, KeyRange{Start: []byte("d"), End: []byte("ee")}, false)
 	require.NoError(t, err)
 
 	// TODO(bilal): Once reading of shared sstables is in, verify that the values
@@ -1582,7 +1589,7 @@ func TestConcurrentExcise(t *testing.T) {
 			require.NoError(t, err)
 			require.NoError(t, w.Close())
 
-			_, err = to.IngestAndExcise([]string{sstPath}, sharedSSTs, nil /* external */, KeyRange{Start: startKey, End: endKey})
+			_, err = to.IngestAndExcise([]string{sstPath}, sharedSSTs, nil, KeyRange{Start: startKey, End: endKey}, false)
 			require.NoError(t, err)
 			return fmt.Sprintf("replicated %d shared SSTs", len(sharedSSTs))
 
@@ -1991,7 +1998,7 @@ func TestIngestExternal(t *testing.T) {
 			)
 			require.NoError(t, err)
 			require.NoError(t, w.Close())
-			_, err = to.IngestAndExcise([]string{sstPath}, nil /* sharedSSTs */, externalFiles, KeyRange{Start: startKey, End: endKey})
+			_, err = to.IngestAndExcise([]string{sstPath}, nil /* shared */, externalFiles, KeyRange{Start: startKey, End: endKey}, false)
 			require.NoError(t, err)
 			return fmt.Sprintf("replicated %d external SSTs", len(externalFiles))
 

metamorphic/ops.go

@@ -923,10 +923,10 @@ func (o *ingestAndExciseOp) run(t *Test, h historyRecorder) {
 
 	if t.testOpts.useExcise {
 		err = firstError(err, t.withRetries(func() error {
-			_, err := t.getDB(o.dbID).IngestAndExcise([]string{path}, nil /* sharedSSTs */, nil /* external */, pebble.KeyRange{
+			_, err := t.getDB(o.dbID).IngestAndExcise([]string{path}, nil /* shared */, nil /* external */, pebble.KeyRange{
 				Start: o.exciseStart,
 				End:   o.exciseEnd,
-			})
+			}, false)
 			return err
 		}))
 	} else {
@@ -1988,7 +1988,7 @@ func (r *replicateOp) runSharedReplicate(
 		return
 	}
 
-	_, err = dest.IngestAndExcise([]string{sstPath}, sharedSSTs, nil /* external */, pebble.KeyRange{Start: r.start, End: r.end})
+	_, err = dest.IngestAndExcise([]string{sstPath}, sharedSSTs, nil /* external */, pebble.KeyRange{Start: r.start, End: r.end}, false)
 	h.Recordf("%s // %v", r, err)
 }
 

open.go

@@ -936,9 +936,7 @@ func (d *DB) replayWAL(
 					panic("pebble: couldn't load all files in WAL entry.")
 				}
 
-				entry, err = d.newIngestedFlushableEntry(
-					meta, seqNum, base.DiskFileNum(ll.Num),
-				)
+				entry, err = d.newIngestedFlushableEntry(meta, seqNum, base.DiskFileNum(ll.Num), KeyRange{})
 				if err != nil {
 					return nil, 0, err
 				}