title | summary | aliases | |
---|---|---|---|
ALTER INDEX |
An overview of the usage of ALTER INDEX for the TiDB database. |
|
The ALTER INDEX
statement is used to modify the visibility of the index to Visible
or Invisible
. Invisible indexes are maintained by DML statements, but will not be used by the query optimizer. This is useful in scenarios where you want to double-check before removing an index permanently. Starting from TiDB v8.0.0, you can make the optimizer select invisible indexes by modifying the system variable tidb_opt_use_invisible_indexes
.
AlterTableStmt
::= 'ALTER' 'IGNORE'? 'TABLE' TableName AlterIndexSpec ( ',' AlterIndexSpec )*
AlterIndexSpec
::= 'ALTER' 'INDEX' Identifier ( 'VISIBLE' | 'INVISIBLE' )
You can modify the visibility of an index using the ALTER TABLE ... ALTER INDEX ...
statement.
{{< copyable "sql" >}}
CREATE TABLE t1 (c1 INT, UNIQUE(c1));
ALTER TABLE t1 ALTER INDEX c1 INVISIBLE;
Query OK, 0 rows affected (0.02 sec)
{{< copyable "sql" >}}
SHOW CREATE TABLE t1;
+-------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Table | Create Table
|
+-------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| t1 | CREATE TABLE `t1` (
`c1` int(11) DEFAULT NULL,
UNIQUE KEY `c1` (`c1`) /*!80000 INVISIBLE */
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_bin |
+-------+------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.00 sec)
The optimizer cannot use the invisible index of c1
.
{{< copyable "sql" >}}
EXPLAIN SELECT c1 FROM t1 ORDER BY c1;
+-------------------------+----------+-----------+---------------+--------------------------------+
| id | estRows | task | access object | operator info |
+-------------------------+----------+-----------+---------------+--------------------------------+
| Sort_4 | 10000.00 | root | | test.t1.c1:asc |
| └─TableReader_8 | 10000.00 | root | | data:TableFullScan_7 |
| └─TableFullScan_7 | 10000.00 | cop[tikv] | table:t1 | keep order:false, stats:pseudo |
+-------------------------+----------+-----------+---------------+--------------------------------+
3 rows in set (0.00 sec)
By comparison, c2
is a visible index and can be used by the optimizer.
{{< copyable "sql" >}}
EXPLAIN SELECT c2 FROM t1 ORDER BY c2;
+------------------------+----------+-----------+------------------------+-------------------------------+
| id | estRows | task | access object | operator info |
+------------------------+----------+-----------+------------------------+-------------------------------+
| IndexReader_13 | 10000.00 | root | | index:IndexFullScan_12 |
| └─IndexFullScan_12 | 10000.00 | cop[tikv] | table:t1, index:c2(c2) | keep order:true, stats:pseudo |
+------------------------+----------+-----------+------------------------+-------------------------------+
2 rows in set (0.00 sec)
Even if you use the USE INDEX
SQL hint to forcibly use indexes, the optimizer still cannot use invisible indexes; otherwise, an error is returned.
{{< copyable "sql" >}}
SELECT * FROM t1 USE INDEX(c1);
ERROR 1176 (42000): Key 'c1' doesn't exist in table 't1'
Note:
"Invisible" here means invisible only to the optimizer. You can still modify or delete invisible indexes.
{{< copyable "sql" >}}
ALTER TABLE t1 DROP INDEX c1;
Query OK, 0 rows affected (0.02 sec)
- Invisible indexes in TiDB are modeled on the equivalent feature from MySQL 8.0.
- Similar to MySQL, TiDB does not permit
PRIMARY KEY
indexes to be made invisible.