MySQL Storage engine

Database storage engine is the underlying software component of database , Database management system uses data engine to create 、 Inquire about 、 Update and delete data operations . Different storage engines provide different storage mechanisms 、 Indexing techniques 、 Lock level and other functions , Specific functions can also be obtained by using different storage engines .

The name storage engine is only used in mysql in .（Oracle There is a corresponding mechanism in , But it's not called a storage engine .Oracle There is no special name in , Namely “ How tables are stored ”）, Different table bottom organization methods ,mysql The default storage engine is InnoDB The way .( The default character set is UTF8)

Complete table building statement
CREATE TABLE t_x (
id int(11) DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

   Be careful ： stay MySQL among , Floating signs can be used for all identifiers (`xxx`) The enclosed . Better not use , Not universal .

   When creating a table, you can specify the storage engine , You can also specify a character set .
   The name storage engine is only used in mysql in .（Oracle There is a corresponding mechanism in , But it's not called a storage engine .Oracle There is no special name in , Namely “ How tables are stored ”）

// View the current mysql Supported storage engines 
show engines;

MySQL5.7 Supported storage engines

Common storage engines

• MyISAM Storage engine

  • The tables it manages have the following characteristics ：

  ​ Use three files to represent each table ：

  ​ Format file — Definition of storage table structure （xxxxx.frm）

  ​ Data files — Store the contents of table rows （xxxxx.MYD）

  ​ Index file — Store indexes on tables （xxxxxx.MYI）

  – agile AUTO_INCREMENT Field handling

  – advantage ： Can be compressed , Save storage space . And can be converted to a read-only table , Improve retrieval efficiency .
  shortcoming ： Unsupported transaction .

• InnoDB Storage engine

  • InnoDB The storage engine is MySQL The default engine .

  • The tables it manages have the following main characteristics ：

  ​ Every InnoDB The table is in the database directory as .frm The format file represents

  ​ InnoDB Table space tablespace Used to store the contents of a table

  ​ Provides a set of log files used to record transactional activities

  ​ use COMMIT( Submit )、SAVEPOINT And ROLLBACK( Roll back ) Support transaction processing

  ​ Provide full ACID compatible

  ​ stay MySQL Provide automatic recovery after server crash

  ​ Many versions （MVCC） And row level locking

  ​ Support the integrity of foreign keys and references , Including cascading deletion and update

  advantage ： Support transactions 、 Row-level locks 、 Foreign keys, etc . The security of this storage engine data is guaranteed .
  The structure of the table is stored in xxx.frm In file
  The data is stored in tablespace In such a table space （ The concept of logic ）, Can't be compressed , Cannot convert to read-only .

• MEMORY Storage engine

  • Use MEMORY The table that stores the engine , Its data is stored in memory , And the length of the line is fixed , These two characteristics make MEMORY The storage engine is very fast .

  • MEMORY The tables managed by the storage engine have the following characteristics ：

  – In the database directory , Each table is marked with .frm Format file representation .

  – Table data and indexes are stored in memory .

  – Table level locking mechanism .

  – Can not contain TEXT or BLOB Field .

  MEMORY The storage engine was formerly known as HEAP engine .

function	MyISAM	MEMORY	InnoDB
Storage limits	256TB	RAM	64TB
Support transactions	NO	NO	YES
Full text index support	YES	NO	NO
Tree index is supported	YES	YES	YES
Support hash index	NO	YES	NO
Support for data caching	NO	NO	YES
Support foreign keys	NO	NO	YES

• MyISAM Table is most suitable for the mixed operation of reading a large amount of data and updating a small amount of data .MyISAM Another use case for tables is to use compressed read-only tables .

• If the query contains more data, the update operation , You should use InnoDB. Its row level locking mechanism and multi version support provide a good concurrency mechanism for the mixed operation of data reading and updating .

• You can use MEMORY Storage engine to store non permanent data , Or data that can be regenerated from disk based tables . Tips ： Which engine to use should be selected flexibly according to the needs , Multiple tables in a database can use different engines to meet various performance and practical requirements . Using the right storage engine will improve the performance of the entire database .