Creating or adding a secondary index

CREATE INDEX name ON table (col_list);

ALTER TABLE tbl_name ADD INDEX name (col_list);

The table remains available for read and write operations while the index is being created. The CREATE INDEX statement only finishes after all transactions that are accessing the table are completed, so that the initial state of the index reflects the most recent contents of the table.

Online DDL support for adding secondary indexes means that you can generally speed the overall process of creating and loading a table and associated indexes by creating the table without secondary indexes, then adding secondary indexes after the data is loaded.

A newly created secondary index contains only the committed data in the table at the time the CREATE INDEX or ALTER TABLE statement finishes executing. It does not contain any uncommitted values, old versions of values, or values marked for deletion but not yet removed from the old index.

Some factors affect the performance, space usage, and semantics of this operation. For details, see Section 15.12.6, “Online DDL Limitations”.

Dropping an index

DROP INDEX name ON table;

ALTER TABLE tbl_name DROP INDEX name;

The table remains available for read and write operations while the index is being dropped. The DROP INDEX statement only finishes after all transactions that are accessing the table are completed, so that the initial state of the index reflects the most recent contents of the table.

Renaming an index

ALTER TABLE tbl_name RENAME INDEX old_index_name TO new_index_name, ALGORITHM=INPLACE, LOCK=NONE;

Adding a FULLTEXT index

CREATE FULLTEXT INDEX name ON table(column);

Adding the first FULLTEXT index rebuilds the table if there is no user-defined FTS_DOC_ID column. Additional FULLTEXT indexes may be added without rebuilding the table.

Adding a SPATIAL index

CREATE TABLE geom (g GEOMETRY NOT NULL);
ALTER TABLE geom ADD SPATIAL INDEX(g), ALGORITHM=INPLACE, LOCK=SHARED;

Changing the index type (USING {BTREE | HASH})

ALTER TABLE tbl_name DROP INDEX i1, ADD INDEX i1(key_part,...) USING BTREE, ALGORITHM=INSTANT;

Adding a primary key

ALTER TABLE tbl_name ADD PRIMARY KEY (column), ALGORITHM=INPLACE, LOCK=NONE;

Rebuilds the table in place. Data is reorganized substantially, making it an expensive operation. ALGORITHM=INPLACE is not permitted under certain conditions if columns have to be converted to NOT NULL.

Restructuring the clustered index always requires copying of table data. Thus, it is best to define the primary key when you create a table, rather than issuing ALTER TABLE ... ADD PRIMARY KEY later.

When you create a UNIQUE or PRIMARY KEY index, MySQL must do some extra work. For UNIQUE indexes, MySQL checks that the table contains no duplicate values for the key. For a PRIMARY KEY index, MySQL also checks that none of the PRIMARY KEY columns contains a NULL.

When you add a primary key using the ALGORITHM=COPY clause, MySQL converts NULL values in the associated columns to default values: 0 for numbers, an empty string for character-based columns and BLOBs, and 0000-00-00 00:00:00 for DATETIME. This is a non-standard behavior that Oracle recommends you not rely on. Adding a primary key using ALGORITHM=INPLACE is only permitted when the SQL_MODE setting includes the strict_trans_tables or strict_all_tables flags; when the SQL_MODE setting is strict, ALGORITHM=INPLACE is permitted, but the statement can still fail if the requested primary key columns contain NULL values. The ALGORITHM=INPLACE behavior is more standard-compliant.

If you create a table without a primary key, InnoDB chooses one for you, which can be the first UNIQUE key defined on NOT NULL columns, or a system-generated key. To avoid uncertainty and the potential space requirement for an extra hidden column, specify the PRIMARY KEY clause as part of the CREATE TABLE statement.

MySQL creates a new clustered index by copying the existing data from the original table to a temporary table that has the desired index structure. Once the data is completely copied to the temporary table, the original table is renamed with a different temporary table name. The temporary table comprising the new clustered index is renamed with the name of the original table, and the original table is dropped from the database.

The online performance enhancements that apply to operations on secondary indexes do not apply to the primary key index. The rows of an InnoDB table are stored in a clustered index organized based on the primary key, forming what some database systems call an “index-organized table”. Because the table structure is closely tied to the primary key, redefining the primary key still requires copying the data.

When an operation on the primary key uses ALGORITHM=INPLACE, even though the data is still copied, it is more efficient than using ALGORITHM=COPY because:

Dropping a primary key

ALTER TABLE tbl_name DROP PRIMARY KEY, ALGORITHM=COPY;

Only ALGORITHM=COPY supports dropping a primary key without adding a new one in the same ALTER TABLE statement.

Dropping a primary key and adding another

ALTER TABLE tbl_name DROP PRIMARY KEY, ADD PRIMARY KEY (column), ALGORITHM=INPLACE, LOCK=NONE;

Data is reorganized substantially, making it an expensive operation.

Adding a column

ALTER TABLE tbl_name ADD COLUMN column_name column_definition, ALGORITHM=INSTANT;

The following limitations apply when the INSTANT algorithm is used to add a column:

Multiple columns may be added in the same ALTER TABLE statement. For example:

ALTER TABLE t1 ADD COLUMN c2 INT, ADD COLUMN c3 INT, ALGORITHM=INSTANT;

INFORMATION_SCHEMA.INNODB_TABLES and INFORMATION_SCHEMA.INNODB_COLUMNS provide metadata for instantly added columns. INFORMATION_SCHEMA.INNODB_TABLES.INSTANT_COLS shows number of columns in the table prior to adding the first instant column. INFORMATION_SCHEMA.INNODB_COLUMNS.HAS_DEFAULT and DEFAULT_VALUE provide metadata about default values for instantly added columns.

Concurrent DML is not permitted when adding an auto-increment column. Data is reorganized substantially, making it an expensive operation. At a minimum, ALGORITHM=INPLACE, LOCK=SHARED is required.

The table is rebuilt if ALGORITHM=INPLACE is used to add a column.

Dropping a column

ALTER TABLE tbl_name DROP COLUMN column_name, ALGORITHM=INPLACE, LOCK=NONE;

Data is reorganized substantially, making it an expensive operation.

Renaming a column

ALTER TABLE tbl CHANGE old_col_name new_col_name data_type, ALGORITHM=INPLACE, LOCK=NONE;

To permit concurrent DML, keep the same data type and only change the column name.

When you keep the same data type and [NOT] NULL attribute, only changing the column name, the operation can always be performed online.

You can also rename a column that is part of a foreign key constraint. The foreign key definition is automatically updated to use the new column name. Renaming a column participating in a foreign key only works with ALGORITHM=INPLACE. If you use the ALGORITHM=COPY clause, or some other condition causes the command to use ALGORITHM=COPY behind the scenes, the ALTER TABLE statement fails.

ALGORITHM=INPLACE is not supported for renaming a generated column.

Reordering columns

To reorder columns, use FIRST or AFTER in CHANGE or MODIFY operations.

ALTER TABLE tbl_name MODIFY COLUMN col_name column_definition FIRST, ALGORITHM=INPLACE, LOCK=NONE;

Data is reorganized substantially, making it an expensive operation.

Changing the column data type

ALTER TABLE tbl_name CHANGE c1 c1 BIGINT, ALGORITHM=COPY;

Changing the column data type is only supported with ALGORITHM=COPY.

Extending VARCHAR column size

ALTER TABLE tbl_name CHANGE COLUMN c1 c1 VARCHAR(255), ALGORITHM=INPLACE, LOCK=NONE;

The number of length bytes required by a VARCHAR column must remain the same. For VARCHAR columns of 0 to 255 bytes in size, one length byte is required to encode the value. For VARCHAR columns of 256 bytes in size or more, two length bytes are required. As a result, in-place ALTER TABLE only supports increasing VARCHAR column size from 0 to 255 bytes, or from 256 bytes to a greater size. In-place ALTER TABLE does not support increasing the size of a VARCHAR column from less than 256 bytes to a size equal to or greater than 256 bytes. In this case, the number of required length bytes changes from 1 to 2, which is only supported by a table copy (ALGORITHM=COPY). For example, attempting to change VARCHAR column size for a single byte character set from VARCHAR(255) to VARCHAR(256) using in-place ALTER TABLE returns this error:

ALTER TABLE tbl_name ALGORITHM=INPLACE, CHANGE COLUMN c1 c1 VARCHAR(256);
ERROR 0A000: ALGORITHM=INPLACE is not supported. Reason: Cannot change
column type INPLACE. Try ALGORITHM=COPY.

Decreasing VARCHAR size using in-place ALTER TABLE is not supported. Decreasing VARCHAR size requires a table copy (ALGORITHM=COPY).

Setting a column default value

ALTER TABLE tbl_name ALTER COLUMN col SET DEFAULT literal, ALGORITHM=INSTANT;

Only modifies table metadata. Default column values are stored in the data dictionary.

Dropping a column default value

ALTER TABLE tbl ALTER COLUMN col DROP DEFAULT, ALGORITHM=INSTANT;

Changing the auto-increment value

ALTER TABLE table AUTO_INCREMENT=next_value, ALGORITHM=INPLACE, LOCK=NONE;

Modifies a value stored in memory, not the data file.

In a distributed system using replication or sharding, you sometimes reset the auto-increment counter for a table to a specific value. The next row inserted into the table uses the specified value for its auto-increment column. You might also use this technique in a data warehousing environment where you periodically empty all the tables and reload them, and restart the auto-increment sequence from 1.

Making a column NULL

ALTER TABLE tbl_name MODIFY COLUMN column_name data_type NULL, ALGORITHM=INPLACE, LOCK=NONE;

Rebuilds the table in place. Data is reorganized substantially, making it an expensive operation.

Making a column NOT NULL

ALTER TABLE tbl_name MODIFY COLUMN column_name data_type NOT NULL, ALGORITHM=INPLACE, LOCK=NONE;

Rebuilds the table in place. STRICT_ALL_TABLES or STRICT_TRANS_TABLES SQL_MODE is required for the operation to succeed. The operation fails if the column contains NULL values. The server prohibits changes to foreign key columns that have the potential to cause loss of referential integrity. See Section 13.1.9, “ALTER TABLE Statement”. Data is reorganized substantially, making it an expensive operation.

Modifying the definition of an ENUM or SET column

CREATE TABLE t1 (c1 ENUM('a', 'b', 'c'));
ALTER TABLE t1 MODIFY COLUMN c1 ENUM('a', 'b', 'c', 'd'), ALGORITHM=INSTANT;

Modifying the definition of an ENUM or SET column by adding new enumeration or set members to the end of the list of valid member values may be performed instantly or in place, as long as the storage size of the data type does not change. For example, adding a member to a SET column that has 8 members changes the required storage per value from 1 byte to 2 bytes; this requires a table copy. Adding members in the middle of the list causes renumbering of existing members, which requires a table copy.

Adding a STORED column

ALTER TABLE t1 ADD COLUMN (c2 INT GENERATED ALWAYS AS (c1 + 1) STORED), ALGORITHM=COPY;

ADD COLUMN is not an in-place operation for stored columns (done without using a temporary table) because the expression must be evaluated by the server.

Modifying STORED column order

ALTER TABLE t1 MODIFY COLUMN c2 INT GENERATED ALWAYS AS (c1 + 1) STORED FIRST, ALGORITHM=COPY;

Rebuilds the table in place.

Dropping a STORED column

ALTER TABLE t1 DROP COLUMN c2, ALGORITHM=INPLACE, LOCK=NONE;

Rebuilds the table in place.

Adding a VIRTUAL column

ALTER TABLE t1 ADD COLUMN (c2 INT GENERATED ALWAYS AS (c1 + 1) VIRTUAL), ALGORITHM=INSTANT;

Adding a virtual column can be performed instantly or in place for non-partitioned tables.

Adding a VIRTUAL is not an in-place operation for partitioned tables.

Modifying VIRTUAL column order

ALTER TABLE t1 MODIFY COLUMN c2 INT GENERATED ALWAYS AS (c1 + 1) VIRTUAL FIRST, ALGORITHM=COPY;

Dropping a VIRTUAL column

ALTER TABLE t1 DROP COLUMN c2, ALGORITHM=INSTANT;

Dropping a VIRTUAL column can be performed instantly or in place for non-partitioned tables.

Adding a foreign key constraint

The INPLACE algorithm is supported when foreign_key_checks is disabled. Otherwise, only the COPY algorithm is supported.

ALTER TABLE tbl1 ADD CONSTRAINT fk_name FOREIGN KEY index (col1)
  REFERENCES tbl2(col2) referential_actions;

Dropping a foreign key constraint

ALTER TABLE tbl DROP FOREIGN KEY fk_name;

Dropping a foreign key can be performed online with the foreign_key_checks option enabled or disabled.

If you do not know the names of the foreign key constraints on a particular table, issue the following statement and find the constraint name in the CONSTRAINT clause for each foreign key:

SHOW CREATE TABLE table\G

Or, query the INFORMATION_SCHEMA.TABLE_CONSTRAINTS table and use the CONSTRAINT_NAME and CONSTRAINT_TYPE columns to identify the foreign key names.

You can also drop a foreign key and its associated index in a single statement:

ALTER TABLE table DROP FOREIGN KEY constraint, DROP INDEX index;

Changing the ROW_FORMAT

ALTER TABLE tbl_name ROW_FORMAT = row_format, ALGORITHM=INPLACE, LOCK=NONE;

Data is reorganized substantially, making it an expensive operation.

For additional information about the ROW_FORMAT option, see Table Options.

Changing the KEY_BLOCK_SIZE

ALTER TABLE tbl_name KEY_BLOCK_SIZE = value, ALGORITHM=INPLACE, LOCK=NONE;

Data is reorganized substantially, making it an expensive operation.

For additional information about the KEY_BLOCK_SIZE option, see Table Options.

Setting persistent table statistics options

ALTER TABLE tbl_name STATS_PERSISTENT=0, STATS_SAMPLE_PAGES=20, STATS_AUTO_RECALC=1, ALGORITHM=INPLACE, LOCK=NONE;

Only modifies table metadata.

Persistent statistics include STATS_PERSISTENT, STATS_AUTO_RECALC, and STATS_SAMPLE_PAGES. For more information, see Section 15.8.10.1, “Configuring Persistent Optimizer Statistics Parameters”.

Specifying a character set

ALTER TABLE tbl_name CHARACTER SET = charset_name, ALGORITHM=INPLACE, LOCK=NONE;

Rebuilds the table if the new character encoding is different.

Converting a character set

ALTER TABLE tbl_name CONVERT TO CHARACTER SET charset_name, ALGORITHM=COPY;

Rebuilds the table if the new character encoding is different.

Optimizing a table

OPTIMIZE TABLE tbl_name;

In-place operation is not supported for tables with FULLTEXT indexes. The operation uses the INPLACE algorithm, but ALGORITHM and LOCK syntax is not permitted.

Rebuilding a table with the FORCE option

ALTER TABLE tbl_name FORCE, ALGORITHM=INPLACE, LOCK=NONE;

Uses ALGORITHM=INPLACE as of MySQL 5.6.17. ALGORITHM=INPLACE is not supported for tables with FULLTEXT indexes.

Performing a "null" rebuild

ALTER TABLE tbl_name ENGINE=InnoDB, ALGORITHM=INPLACE, LOCK=NONE;

Uses ALGORITHM=INPLACE as of MySQL 5.6.17. ALGORITHM=INPLACE is not supported for tables with FULLTEXT indexes.

Renaming a table

ALTER TABLE old_tbl_name RENAME TO new_tbl_name, ALGORITHM=INSTANT;

Renaming a table can be performed instantly or in place. MySQL renames files that correspond to the table tbl_name without making a copy. (You can also use the RENAME TABLE statement to rename tables. See Section 13.1.36, “RENAME TABLE Statement”.) Privileges granted specifically for the renamed table are not migrated to the new name. They must be changed manually.

Renaming a general tablespace

ALTER TABLESPACE tablespace_name RENAME TO new_tablespace_name;

ALTER TABLESPACE ... RENAME TO uses the INPLACE algorithm but does not support the ALGORITHM clause.

Enabling or disabling general tablespace encryption

ALTER TABLESPACE tablespace_name ENCRYPTION='Y';

ALTER TABLESPACE ... ENCRYPTION uses the INPLACE algorithm but does not support the ALGORITHM clause.

For related information, see Section 15.13, “InnoDB Data-at-Rest Encryption”.

Enabling or disabling file-per-table tablespace encryption

ALTER TABLE tbl_name ENCRYPTION='Y', ALGORITHM=COPY;

For related information, see Section 15.13, “InnoDB Data-at-Rest Encryption”.