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

Jump to bottom

sinkv2 (ticdc): limit batch dml size #7960

Merged
merged 7 commits into from
Dec 26, 2022
Merged

sinkv2 (ticdc): limit batch dml size #7960

Show file tree
Hide file tree
Changes from 4 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
5528c34
limit batch dml size
asddongmen Dec 22, 2022
5fe81c5
Merge branch 'master' into fix-batch-dml-error
Rustin170506 Dec 26, 2022
fe8f888
add unit test
asddongmen Dec 26, 2022
1aba395
Merge branch 'fix-batch-dml-error' of github.com:asddongmen/ticdc int…
asddongmen Dec 26, 2022
a7c087a
address comments
asddongmen Dec 26, 2022
a92c641
address comments
asddongmen Dec 26, 2022
047fc14
Merge branch 'master' into fix-batch-dml-error
ti-chi-bot Dec 26, 2022
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
115 changes: 82 additions & 33 deletions cdc/sinkv2/eventsink/txn/mysql/mysql.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -273,72 +273,121 @@ func convertBinaryToString(row *model.RowChangedEvent) {
}
}

// TODO: Find a way to make batch delete dmls more efficient.
func groupRowsByType(
func (s *mysqlBackend) groupRowsByType(
event *eventsink.TxnCallbackableEvent,
tableInfo *timodel.TableInfo,
spiltUpdate bool,
) (insertRows, updateRows, deleteRows []*sqlmodel.RowChange) {
) (insertRows, updateRows, deleteRows [][]*sqlmodel.RowChange) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

could you add a test case to verify the issue.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

sure, I will do it.

preAllocateSize := len(event.Event.Rows)
if preAllocateSize > s.cfg.MaxTxnRow {
preAllocateSize = s.cfg.MaxTxnRow
}

insertRow := make([]*sqlmodel.RowChange, 0, preAllocateSize)
updateRow := make([]*sqlmodel.RowChange, 0, preAllocateSize)
deleteRow := make([]*sqlmodel.RowChange, 0, preAllocateSize)

for _, row := range event.Event.Rows {
convertBinaryToString(row)

if row.IsInsert() {
insertRows = append(
insertRows,
insertRow = append(
insertRow,
convert2RowChanges(row, tableInfo, sqlmodel.RowChangeInsert))
} else if row.IsDelete() {
deleteRows = append(
deleteRows,
if len(insertRow) >= s.cfg.MaxTxnRow {
insertRows = append(insertRows, insertRow)
insertRow = make([]*sqlmodel.RowChange, 0, preAllocateSize)
}
}

if row.IsDelete() {
deleteRow = append(
deleteRow,
convert2RowChanges(row, tableInfo, sqlmodel.RowChangeDelete))
} else if row.IsUpdate() {
if len(deleteRow) >= s.cfg.MaxTxnRow {
deleteRows = append(deleteRows, deleteRow)
deleteRow = make([]*sqlmodel.RowChange, 0, preAllocateSize)
}
}

if row.IsUpdate() {
if spiltUpdate {
deleteRows = append(
deleteRows,
deleteRow = append(
deleteRow,
convert2RowChanges(row, tableInfo, sqlmodel.RowChangeDelete))
insertRows = append(
insertRows,
if len(deleteRow) >= s.cfg.MaxTxnRow {
deleteRows = append(deleteRows, deleteRow)
deleteRow = make([]*sqlmodel.RowChange, 0, s.cfg.MaxTxnRow)
}
insertRow = append(
insertRow,
convert2RowChanges(row, tableInfo, sqlmodel.RowChangeInsert))
if len(insertRow) >= s.cfg.MaxTxnRow {
insertRows = append(insertRows, insertRow)
insertRow = make([]*sqlmodel.RowChange, 0, s.cfg.MaxTxnRow)
}
} else {
updateRows = append(
updateRows,
updateRow = append(
updateRow,
convert2RowChanges(row, tableInfo, sqlmodel.RowChangeUpdate))
if len(updateRow) >= s.cfg.MaxTxnRow {
updateRows = append(updateRows, updateRow)
updateRow = make([]*sqlmodel.RowChange, 0, s.cfg.MaxTxnRow)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why do we use s.cfg.MaxTxnRow here but use preAllocateSize above?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In the most time len(event) is far more less than s.cfg.MaxTxnRow, it will be a waste if just use s.cfg.MaxTxnRow.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you for your reminder.

}
}
}
}

if len(insertRow) > 0 {
insertRows = append(insertRows, insertRow)
}
if len(updateRow) > 0 {
updateRows = append(updateRows, updateRow)
}
if len(deleteRow) > 0 {
deleteRows = append(deleteRows, deleteRow)
}

return
}

func batchSingleTxnDmls(
func (s *mysqlBackend) batchSingleTxnDmls(
event *eventsink.TxnCallbackableEvent,
tableInfo *timodel.TableInfo,
translateToInsert bool,
) (sqls []string, values [][]interface{}) {
insertRows, updateRows, deleteRows := groupRowsByType(event, tableInfo, !translateToInsert)
insertRows, updateRows, deleteRows := s.groupRowsByType(event, tableInfo, !translateToInsert)

if len(deleteRows) > 0 {
sql, value := sqlmodel.GenDeleteSQL(deleteRows...)
sqls = append(sqls, sql)
values = append(values, value)
for _, rows := range deleteRows {
sql, value := sqlmodel.GenDeleteSQL(rows...)
sqls = append(sqls, sql)
values = append(values, value)
}
}

// handle insert
if len(insertRows) > 0 {
if translateToInsert {
sql, value := sqlmodel.GenInsertSQL(sqlmodel.DMLInsert, insertRows...)
sqls = append(sqls, sql)
values = append(values, value)
} else {
sql, value := sqlmodel.GenInsertSQL(sqlmodel.DMLReplace, insertRows...)
sqls = append(sqls, sql)
values = append(values, value)
for _, rows := range insertRows {
if translateToInsert {
sql, value := sqlmodel.GenInsertSQL(sqlmodel.DMLInsert, rows...)
sqls = append(sqls, sql)
values = append(values, value)
} else {
sql, value := sqlmodel.GenInsertSQL(sqlmodel.DMLReplace, rows...)
sqls = append(sqls, sql)
values = append(values, value)
}
}
}

// handle update
if len(updateRows) > 0 {
// TODO: do a testing on update performance.
sql, value := sqlmodel.GenUpdateSQL(updateRows...)
sqls = append(sqls, sql)
values = append(values, value)
for _, rows := range updateRows {
sql, value := sqlmodel.GenUpdateSQL(rows...)
sqls = append(sqls, sql)
values = append(values, value)
}
}

return
Expand Down Expand Up @@ -417,7 +466,7 @@ func (s *mysqlBackend) prepareDMLs() *preparedDMLs {
if hasHandleKey(tableColumns) {
// TODO(dongmen): find a better way to get table info.
tableInfo := model.BuildTiDBTableInfo(tableColumns, firstRow.IndexColumns)
sql, value := batchSingleTxnDmls(event, tableInfo, translateToInsert)
sql, value := s.batchSingleTxnDmls(event, tableInfo, translateToInsert)
sqls = append(sqls, sql...)
values = append(values, value...)
continue
Expand Down
231 changes: 231 additions & 0 deletions cdc/sinkv2/eventsink/txn/mysql/mysql_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1457,3 +1457,234 @@ func TestNetworkPartition(t *testing.T) {
err := ms.execDMLWithMaxRetries(ctx, &preparedDMLs{})
require.Equal(t, context.Canceled, err)
}

func TestGroupRowsByType(t *testing.T) {
ctx := context.Background()
ms := newMySQLBackendWithoutDB(ctx)
testCases := []struct {
input []*model.RowChangedEvent
maxTxnRow int
}{
{ // delete event
input: []*model.RowChangedEvent{
{
StartTs: 418658114257813514,
CommitTs: 418658114257813515,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
PreColumns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.UniqueKeyFlag,
Value: 1,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.UniqueKeyFlag,
Value: 1,
}},
IndexColumns: [][]int{{1, 2}},
},
{
StartTs: 418658114257813514,
CommitTs: 418658114257813515,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
PreColumns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.UniqueKeyFlag,
Value: 2,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.UniqueKeyFlag,
Value: 2,
}},
IndexColumns: [][]int{{1, 2}},
},
{
StartTs: 418658114257813514,
CommitTs: 418658114257813515,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
PreColumns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.UniqueKeyFlag,
Value: 2,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.UniqueKeyFlag,
Value: 2,
}},
IndexColumns: [][]int{{1, 2}},
},
{
StartTs: 418658114257813514,
CommitTs: 418658114257813515,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
PreColumns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.UniqueKeyFlag,
Value: 2,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.UniqueKeyFlag,
Value: 2,
}},
IndexColumns: [][]int{{1, 2}},
},
},
maxTxnRow: 2,
},
{ // insert event
input: []*model.RowChangedEvent{
{
StartTs: 418658114257813516,
CommitTs: 418658114257813517,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
Columns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag | model.HandleKeyFlag,
Value: 1,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag | model.HandleKeyFlag,
Value: 1,
}},
IndexColumns: [][]int{{1, 1}},
},
{
StartTs: 418658114257813516,
CommitTs: 418658114257813517,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
Columns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}},
IndexColumns: [][]int{{2, 2}},
},
{
StartTs: 418658114257813516,
CommitTs: 418658114257813517,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
Columns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}},
IndexColumns: [][]int{{2, 2}},
},
{
StartTs: 418658114257813516,
CommitTs: 418658114257813517,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
Columns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}},
IndexColumns: [][]int{{2, 2}},
},

{
StartTs: 418658114257813516,
CommitTs: 418658114257813517,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
Columns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}},
IndexColumns: [][]int{{2, 2}},
},

{
StartTs: 418658114257813516,
CommitTs: 418658114257813517,
Table: &model.TableName{Schema: "common_1", Table: "uk_without_pk"},
Columns: []*model.Column{nil, {
Name: "a1",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}, {
Name: "a3",
Type: mysql.TypeLong,
Flag: model.BinaryFlag | model.MultipleKeyFlag |
model.HandleKeyFlag | model.HandleKeyFlag,
Value: 2,
}},
IndexColumns: [][]int{{2, 2}},
},
},
maxTxnRow: 4,
},
}
for _, tc := range testCases {
event := &eventsink.TxnCallbackableEvent{
Event: &model.SingleTableTxn{Rows: testCases[0].input},
}
colums := tc.input[0].Columns
if len(colums) == 0 {
colums = tc.input[0].PreColumns
}
tableInfo := model.BuildTiDBTableInfo(colums, tc.input[0].IndexColumns)
ms.cfg.MaxTxnRow = tc.maxTxnRow
inserts, updates, deletes := ms.groupRowsByType(event, tableInfo, false)
for _, rows := range inserts {
require.LessOrEqual(t, len(rows), tc.maxTxnRow)
}
for _, rows := range updates {
require.LessOrEqual(t, len(rows), tc.maxTxnRow)
}
for _, rows := range deletes {
require.LessOrEqual(t, len(rows), tc.maxTxnRow)
}
}
}