feat: merge-DML batching optimization for binlog apply

doc/command-line-flags.md

@@ -158,6 +158,21 @@ While the ongoing estimated number of rows is still heuristic, it's almost exact
 
 Without this parameter, migration is a _noop_: testing table creation and validity of migration, but not touching data.
 
+### is-merge-dml-event
+
+When enabled, batched binlog DML events are merged in memory before applying them to the ghost table. Only effective when the migration unique key uses numeric column types (`int`, `bigint`, `decimal`, `float`, etc.).
+
+**Batching:** All DML events in a batch are grouped by type — inserts and updates become a single multi-row `REPLACE INTO`, deletes become a single `DELETE WHERE (pk) IN (...)`.
+
+**Deduplication:** Repeated changes to the same unique key within a batch collapse to the final state (last writer wins).
+
+**Range filtering:** When a binlog event's unique key value is beyond `MigrationIterationRangeMaxValues` but within `MigrationRangeMaxValues`, the event is skipped — that data will be synced by the row-copy chunk. Events beyond `MigrationRangeMaxValues` or below `MigrationIterationRangeMaxValues` are applied normally.
+
+Automatically disabled when:
+- The chosen unique key contains non-numeric columns (TEXT, BLOB, JSON, etc.)
+- The chosen unique key has nullable columns (NULL breaks comparison and dedup semantics)
+- The table has multiple unique indexes (REPLACE semantics are unsafe with secondary unique constraints)
+
 ### force-named-cut-over
 
 If given, a `cut-over` command must name the migrated table, or else ignored.

go/base/context.go

@@ -211,7 +211,9 @@ type MigrationContext struct {
 	controlReplicasLagResult               mysql.ReplicationLagResult
 	TotalRowsCopied                        int64
 	TotalDMLEventsApplied                  int64
+	TotalDMLEventsIgnored                  int64
 	DMLBatchSize                           int64
+	IsMergeDMLEvents                       bool
 	isThrottled                            bool
 	throttleReason                         string
 	throttleReasonHint                     ThrottleReasonHint

go/cmd/gh-ost/main.go

@@ -108,6 +108,7 @@ func main() {
 	exponentialBackoffMaxInterval := flag.Int64("exponential-backoff-max-interval", 64, "Maximum number of seconds to wait between attempts when performing various operations with exponential backoff.")
 	chunkSize := flag.Int64("chunk-size", 1000, "amount of rows to handle in each iteration (allowed range: 10-100,000)")
 	dmlBatchSize := flag.Int64("dml-batch-size", 10, "batch size for DML events to apply in a single transaction (range 1-1000)")
+	flag.BoolVar(&migrationContext.IsMergeDMLEvents, "is-merge-dml-event", false, "Merge DML Binlog Event")
 	defaultRetries := flag.Int64("default-retries", 60, "Default number of retries for various operations before panicking")
 	flag.BoolVar(&migrationContext.PanicOnWarnings, "panic-on-warnings", false, "Panic when SQL warnings are encountered when copying a batch indicating data loss")
 	cutOverLockTimeoutSeconds := flag.Int64("cut-over-lock-timeout-seconds", 3, "Max number of seconds to hold locks on tables while attempting to cut-over (retry attempted when lock exceeds timeout) or attempting instant DDL")