Use a pool for document comparisons (#169)

Rather than allocating a fresh buffer for each document, this changeset uses a dynamic buffer pool. This relieves GC pressure when comparing documents, which is particularly relevant during generation 0 (i.e., initial comparison of all documents).

Author: FGasper

go.mod

@@ -44,6 +44,7 @@ require (
 	github.com/klauspost/compress v1.16.7 // indirect
 	github.com/klauspost/cpuid/v2 v2.2.10 // indirect
 	github.com/leodido/go-urn v1.4.0 // indirect
+	github.com/libp2p/go-buffer-pool v0.1.0 // indirect
 	github.com/mattn/go-colorable v0.1.13 // indirect
 	github.com/mattn/go-isatty v0.0.20 // indirect
 	github.com/mattn/go-runewidth v0.0.9 // indirect

go.sum

@@ -65,6 +65,8 @@ github.com/klauspost/cpuid/v2 v2.2.10/go.mod h1:hqwkgyIinND0mEev00jJYCxPNVRVXFQe
 github.com/knz/go-libedit v1.10.1/go.mod h1:MZTVkCWyz0oBc7JOWP3wNAzd002ZbM/5hgShxwh4x8M=
 github.com/leodido/go-urn v1.4.0 h1:WT9HwE9SGECu3lg4d/dIA+jxlljEa1/ffXKmRjqdmIQ=
 github.com/leodido/go-urn v1.4.0/go.mod h1:bvxc+MVxLKB4z00jd1z+Dvzr47oO32F/QSNjSBOlFxI=
+github.com/libp2p/go-buffer-pool v0.1.0 h1:oK4mSFcQz7cTQIfqbe4MIj9gLW+mnanjyFtc6cdF0Y8=
+github.com/libp2p/go-buffer-pool v0.1.0/go.mod h1:N+vh8gMqimBzdKkSMVuydVDq+UV5QTWy5HSiZacSbPg=
 github.com/mattn/go-colorable v0.1.13 h1:fFA4WZxdEF4tXPZVKMLwD8oUnCTTo08duU7wxecdEvA=
 github.com/mattn/go-colorable v0.1.13/go.mod h1:7S9/ev0klgBDR4GtXTXX8a3vIGJpMovkB8vQcUbaXHg=
 github.com/mattn/go-isatty v0.0.16/go.mod h1:kYGgaQfpe5nmfYZH+SKPsOc2e4SrIfOl2e/yFXSvRLM=

internal/verifier/compare.go

@@ -13,6 +13,7 @@ import (
 	"github.com/10gen/migration-verifier/internal/types"
 	"github.com/10gen/migration-verifier/internal/util"
 	"github.com/10gen/migration-verifier/option"
+	pool "github.com/libp2p/go-buffer-pool"
 	"github.com/pkg/errors"
 	"go.mongodb.org/mongo-driver/v2/bson"
 	"go.mongodb.org/mongo-driver/v2/mongo"
@@ -179,6 +180,9 @@ func (verifier *Verifier) compareDocsFromChannels(
 			return errors.Wrap(err, "failed to compare documents")
 		}
 
+		pool.Put(srcDoc.doc)
+		pool.Put(dstDoc.doc)
+
 		for i := range mismatches {
 			mismatches[i].SrcTimestamp = option.Some(srcDoc.ts)
 			mismatches[i].DstTimestamp = option.Some(dstDoc.ts)
@@ -326,6 +330,8 @@ func (verifier *Verifier) compareDocsFromChannels(
 				SrcTimestamp: option.Some(docWithTs.ts),
 			},
 		)
+
+		pool.Put(docWithTs.doc)
 	}
 
 	for _, docWithTs := range dstCache {
@@ -343,6 +349,8 @@ func (verifier *Verifier) compareDocsFromChannels(
 				DstTimestamp: option.Some(docWithTs.ts),
 			},
 		)
+
+		pool.Put(docWithTs.doc)
 	}
 
 	return results, srcDocCount, srcByteCount, nil
@@ -352,14 +360,21 @@ func getDocIdFromComparison(
 	docCompareMethod DocCompareMethod,
 	doc bson.Raw,
 ) bson.RawValue {
+	var docID bson.RawValue
+
 	switch docCompareMethod {
 	case DocCompareBinary, DocCompareIgnoreOrder:
-		return doc.Lookup("_id")
+		docID = doc.Lookup("_id")
 	case DocCompareToHashedIndexKey:
-		return doc.Lookup(docKeyInHashedCompare, "_id")
+		docID = doc.Lookup(docKeyInHashedCompare, "_id")
 	default:
 		panic("bad doc compare method: " + docCompareMethod)
 	}
+
+	// We clone the value because the document might be from a pool.
+	docID.Value = slices.Clone(docID.Value)
+
+	return docID
 }
 
 func getDocKeyValues(
@@ -527,11 +542,14 @@ func iterateCursorToChannel(
 			return errors.Wrap(err, "reading cluster time from session")
 		}
 
+		buf := pool.Get(len(cursor.Current))
+		copy(buf, cursor.Current)
+
 		err = chanutil.WriteWithDoneCheck(
 			sctx,
 			writer,
 			docWithTs{
-				doc: slices.Clone(cursor.Current),
+				doc: buf,
 				ts:  clusterTime,
 			},
 		)