The data volume of some tables grows very fast, and the corresponding SQL scans a lot of invalid data, causing the SQL to slow down. After confirmation, these large tables are all running water, records, and log type data, and only 1 to 3 need to be retained. Months, at this time it is necessary to clean the data in the table to achieve weight reduction.
In this article, I will explain why it is not recommended to delete data from the aspects of InnoDB storage space distribution, the impact of delete on performance, and optimization suggestions.
As you can see from this picture, the InnoDB storage structure mainly consists of two parts: logical storage structure and physical storage structure.
Logically, it is composed of tablespace —> segment or inode —> area Extent —> data page. The logical management unit of Innodb is segment, and the minimum unit of space allocation is extent. Each Each segment will allocate 32 pages from the table space FREE_PAGE. When these 32 pages are not enough, they will be expanded according to the following principles: if the current extent is less than 1 extent, it will be expanded to 1 extent; when the table space is less than 32MB, each segment will be expanded to 1 extent. Expand one extent at a time; if the table space is larger than 32MB, extend 4 extents at a time.
Physically, it mainly consists of system user data files and log files. The data files mainly store MySQL dictionary data and user data. The log files record the change records of the data page and are used for recovery when MySQL crashes.
InnoDB storage includes three types of table spaces: system table space, user table space, and Undo table space.
System table space:Mainly stores MySQL internal data dictionary data, such as data under information_schema.
User table space:When innodb_file_per_table=1 is turned on, the data table is separated from the system table space and stored in the data file with the table_name.ibd command, and the structural information is stored in table_name.frm in the file.
Undo table space:Stores Undo information, such as snapshot consistent read and flashback, which all use undo information.
Starting from MySQL 8.0, users are allowed to customize table spaces. The specific syntax is as follows:
CREATE TABLESPACE tablespace_name ADD DATAFILE 'file_name' #数据文件名 USE LOGFILE GROUP logfile_group #自定义日志文件组，一般每组2个logfile。 [EXTENT_SIZE [=] extent_size] #区大小 [INITIAL_SIZE [=] initial_size] #初始化大小 [AUTOEXTEND_SIZE [=] autoextend_size] #自动扩宽尺寸 [MAX_SIZE [=] max_size] #单个文件最大size，最大是32G。 [NODEGROUP [=] nodegroup_id] #节点组 [WAIT] [COMMENT [=] comment_text] ENGINE [=] engine_name
The advantage of this is that it can separate hot and cold data and use HDD and SSD to store them respectively. It can achieve efficient access to data and save costs. For example, you can add two 500G hard disks, create a volume group vg, divide the logical volume lv, create a data directory and mount the corresponding lv. Assume that the two divided directories are /hot_data and/cold_data.
In this way, core business tables such as user tables and order tables can be stored on high-performance SSD disks, and some logs and flow tables can be stored on ordinary HDDs. The main operation steps are as follows:
#创建热数据表空间 create tablespace tbs_data_hot add datafile '/hot_data/tbs_data_hot01.dbf' max_size 20G; #创建核心业务表存储在热数据表空间 create table booking(id bigint not null primary key auto_increment, …… ) tablespace tbs_data_hot; #创建冷数据表空间 create tablespace tbs_data_cold add datafile '/hot_data/tbs_data_cold01.dbf' max_size 20G; #创建日志，流水，备份类的表存储在冷数据表空间 create table payment_log(id bigint not null primary key auto_increment, …… ) tablespace tbs_data_cold; #可以移动表到另一个表空间 alter table payment_log tablespace tbs_data_hot;
mysql> create table user(id bigint not null primary key auto_increment, -> name varchar(20) not null default '' comment '姓名', -> age tinyint not null default 0 comment 'age', -> gender char(1) not null default 'M' comment '性别', -> phone varchar(16) not null default '' comment '手机号', -> create_time datetime NOT NULL DEFAULT CURRENT_TIMESTAMP COMMENT '创建时间', -> update_time datetime NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT '修改时间' -> ) engine = InnoDB DEFAULT CHARSET=utf8mb4 COMMENT '用户信息表'; Query OK, 0 rows affected (0.26 sec)
# ls -lh user1.ibd -rw-r----- 1 mysql mysql 96K Nov 6 12:48 user.ibd
When setting the parameter innodb_file_per_table=1, a segment will be automatically created when creating the table, and an extent will be allocated at the same time, containing 32 data page to store data. The default size of the empty table created in this way is 96KB. After the extent is used, 64 connection pages will be applied for. In this way, for some small tables or undo segments, less space can be applied for at the beginning to save disk capacity. s expenses.
# python2 py_innodb_page_info.py -v /data2/mysql/test/user.ibd page offset 00000000, page typepage offset 00000001, page type page offset 00000002, page type page offset 00000003, page type , page level <0000> page offset 00000000, page type page offset 00000000, page type Total number of page: 6: #总共分配的页数 Freshly Allocated Page: 2 #可用的数据页 Insert Buffer Bitmap: 1 #插入缓冲页 File Space Header: 1 #文件空间头 B-tree Node: 1 #数据页 File Segment inode: 1 #文件端inonde，如果是在ibdata1.ibd上会有多个inode。
mysql> DELIMITER $$ mysql> CREATE PROCEDURE insert_user_data(num INTEGER) -> BEGIN -> DECLARE v_i int unsigned DEFAULT 0; -> set autocommit= 0; -> WHILE v_i < num DO -> insert into user(`name`, age, gender, phone) values (CONCAT('lyn',v_i), mod(v_i,120), 'M', CONCAT('152',ROUND(RAND(1)*100000000))); -> SET v_i = v_i+1; -> END WHILE; -> commit; -> END $$ Query OK, 0 rows affected (0.01 sec) mysql> DELIMITER ; #插入10w数据 mysql> call insert_user_data(100000); Query OK, 0 rows affected (6.69 sec)
# ls -lh user.ibd -rw-r----- 1 mysql mysql 14M Nov 6 10:58 /data2/mysql/test/user.ibd # python2 py_innodb_page_info.py -v /data2/mysql/test/user.ibd page offset 00000000, page typepage offset 00000001, page type page offset 00000002, page type page offset 00000003, page type , page level <0001> #增加了一个非叶子节点，树的高度从1变为2. ........................................................ page offset 00000000, page type Total number of page: 896: Freshly Allocated Page: 493 Insert Buffer Bitmap: 1 File Space Header: 1 B-tree Node: 400 File Segment inode: 1
mysql> select min(id),max(id),count(*) from user; +---------+---------+----------+ | min(id) | max(id) | count(*) | +---------+---------+----------+ | 1 | 100000 | 100000 | +---------+---------+----------+ 1 row in set (0.05 sec) #删除50000条数据，理论上空间应该从14MB变长7MB左右。 mysql> delete from user limit 50000; Query OK, 50000 rows affected (0.25 sec) #数据文件大小依然是14MB，没有缩小。 # ls -lh /data2/mysql/test/user1.ibd -rw-r----- 1 mysql mysql 14M Nov 6 13:22 /data2/mysql/test/user.ibd #数据页没有被回收。 # python2 py_innodb_page_info.py -v /data2/mysql/test/user.ibd page offset 00000000, page typepage offset 00000001, page type page offset 00000002, page type page offset 00000003, page type , page level <0001> ........................................................ page offset 00000000, page type Total number of page: 896: Freshly Allocated Page: 493 Insert Buffer Bitmap: 1 File Space Header: 1 B-tree Node: 400 File Segment inode: 1 #在MySQL内部是标记删除，
mysql> use information_schema; Database changed mysql> SELECT A.SPACE AS TBL_SPACEID, A.TABLE_ID, A.NAME AS TABLE_NAME, FILE_FORMAT, ROW_FORMAT, SPACE_TYPE, B.INDEX_ID , B.NAME AS INDEX_NAME, PAGE_NO, B.TYPE AS INDEX_TYPE FROM INNODB_SYS_TABLES A LEFT JOIN INNODB_SYS_INDEXES B ON A.TABLE_ID =B.TABLE_ID WHERE A.NAME = 'test/user1'; +-------------+----------+------------+-------------+------------+------------+----------+------------+---------+------------+ | TBL_SPACEID | TABLE_ID | TABLE_NAME | FILE_FORMAT | ROW_FORMAT | SPACE_TYPE | INDEX_ID | INDEX_NAME | PAGE_NO | INDEX_TYPE | +-------------+----------+------------+-------------+------------+------------+----------+------------+---------+------------+ | 1283 | 1207 | test/user | Barracuda | Dynamic | Single | 2236 | PRIMARY | 3 | 3 | +-------------+----------+------------+-------------+------------+------------+----------+------------+---------+------------+ 1 row in set (0.01 sec) PAGE_NO = 3 标识B-tree的root page是3号页，INDEX_TYPE = 3是聚集索引。 INDEX_TYPE取值如下： 0 = nonunique secondary index; 1 = automatically generated clustered index (GEN_CLUST_INDEX); 2 = unique nonclustered index; 3 = clustered index; 32 = full-text index; #收缩空间再后进行观察
MySQL will not actually delete the space internally, and will mark the deletion, that is, delflag:N Modify it to delflag:Y. After commit, it will be purged into the delete linked list. If a larger record is inserted next time, the space after delete will not be reused. If the inserted record is less than or equal to the deleted record, the space will be reused. This block The content can be analyzed through Zhishutang’s innblock tool.
We know that data is stored on the file system, and 100% of the physical space allocated to it cannot always be used. Delete the data Some "holes" will be left on the page, or random writing (non-linear increase of the clustered index) will cause page splits. Page splits will cause the page utilization space to be less than 50%. In addition, additions, deletions and modifications to the table will cause corresponding secondary changes. Random additions, deletions, and modifications of level-level index values will also cause some "holes" to be left on the data pages in the index structure. Although these holes may be reused, they will eventually lead to part of the physical space being unused, that is, fragmentation. .
At the same time, even if the fill factor is set to 100%, Innodb will actively leave 1/16 of the page page as a reserved space (An innodb_fill_factor setting of 100 leaves 1/16 of the space in clustered index pages free for future index growth) to prevent row overflow caused by update.
mysql> select table_schema, -> table_name,ENGINE, -> round(DATA_LENGTH/1024/1024+ INDEX_LENGTH/1024/1024) total_mb,TABLE_ROWS, -> round(DATA_LENGTH/1024/1024) data_mb, round(INDEX_LENGTH/1024/1024) index_mb, round(DATA_FREE/1024/1024) free_mb, round(DATA_FREE/DATA_LENGTH*100,2) free_ratio -> from information_schema.TABLES where TABLE_SCHEMA= 'test' -> and TABLE_NAME= 'user'; +--------------+------------+--------+----------+------------+---------+----------+---------+------------+ | table_schema | table_name | ENGINE | total_mb | TABLE_ROWS | data_mb | index_mb | free_mb | free_ratio | +--------------+------------+--------+----------+------------+---------+----------+---------+------------+ | test | user | InnoDB | 4 | 50000 | 4 | 0 | 6 | 149.42 | +--------------+------------+--------+----------+------------+---------+----------+---------+------------+ 1 row in set (0.00 sec)
Data_free is the number of unused bytes allocated, which does not mean that it is completely fragmented space.
For InnoDB tables, you can use the following command to recycle fragments and release space. This is a random read IO operation, which will be more time-consuming and will also block normal operations on the table. DML operation also requires more disk space. For RDS, the disk space may be filled up instantly, the instance will be locked instantly, and the application cannot perform DML operations, so execution in the online environment is prohibited.
#执行InnoDB的碎片回收 mysql> alter table user engine=InnoDB; Query OK, 0 rows affected (9.00 sec) Records: 0 Duplicates: 0 Warnings: 0 ##执行完之后，数据文件大小从14MB降低到10M。 # ls -lh /data2/mysql/test/user1.ibd -rw-r----- 1 mysql mysql 10M Nov 6 16:18 /data2/mysql/test/user.ibd
mysql> select table_schema, ->table_name,ENGINE, ->round(DATA_LENGTH/1024/1024+ INDEX_LENGTH/1024/1024) total_mb,TABLE_ROWS, ->round(DATA_LENGTH/1024/1024) data_mb, ->round(INDEX_LENGTH/1024/1024) index_mb, ->round(DATA_FREE/1024/1024) free_mb, ->round(DATA_FREE/DATA_LENGTH*100,2) free_ratio from information_schema.TABLES where TABLE_SCHEMA= 'test' and TABLE_NAME= 'user'; +--------------+------------+--------+----------+------------+---------+----------+---------+------------+ | table_schema | table_name | ENGINE | total_mb | TABLE_ROWS | data_mb | index_mb | free_mb | free_ratio | +--------------+------------+--------+----------+------------+---------+----------+---------+------------+ | test | user | InnoDB | 5 | 50000 | 5 | 0 | 2 | 44.29 | +--------------+------------+--------+----------+------------+---------+----------+---------+------------+ 1 row in set (0.00 sec)
#插入100W数据 mysql> call insert_user_data(1000000); Query OK, 0 rows affected (35.99 sec) #添加相关索引 mysql> alter table user add index idx_name(name), add index idx_phone(phone); Query OK, 0 rows affected (6.00 sec) Records: 0 Duplicates: 0 Warnings: 0 #表上索引统计信息 mysql> show index from user; +-------+------------+-----------+--------------+-------------+-----------+-------------+----------+--------+------+------------+---------+---------------+ | Table | Non_unique | Key_name | Seq_in_index | Column_name | Collation | Cardinality | Sub_part | Packed | Null | Index_type | Comment | Index_comment | +-------+------------+-----------+--------------+-------------+-----------+-------------+----------+--------+------+------------+---------+---------------+ | user | 0 | PRIMARY | 1 | id | A | 996757 | NULL | NULL | | BTREE | | | | user | 1 | idx_name | 1 | name | A | 996757 | NULL | NULL | | BTREE | | | | user | 1 | idx_phone | 1 | phone | A | 2 | NULL | NULL | | BTREE | | | +-------+------------+-----------+--------------+-------------+-----------+-------------+----------+--------+------+------------+---------+---------------+ 3 rows in set (0.00 sec) #重置状态变量计数 mysql> flush status; Query OK, 0 rows affected (0.00 sec) #执行SQL语句 mysql> select id, age ,phone from user where name like 'lyn12%'; +--------+-----+-------------+ | id | age | phone | +--------+-----+-------------+ | 124 | 3 | 15240540354 | | 1231 | 30 | 15240540354 | | 12301 | 60 | 15240540354 | ............................. | 129998 | 37 | 15240540354 | | 129999 | 38 | 15240540354 | | 130000 | 39 | 15240540354 | +--------+-----+-------------+ 11111 rows in set (0.03 sec) mysql> explain select id, age ,phone from user where name like 'lyn12%'; +----+-------------+-------+-------+---------------+----------+---------+------+-------+-----------------------+ | id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra | +----+-------------+-------+-------+---------------+----------+---------+------+-------+-----------------------+ | 1 | SIMPLE | user | range | idx_name | idx_name | 82 | NULL | 22226 | Using index condition | +----+-------------+-------+-------+---------------+----------+---------+------+-------+-----------------------+ 1 row in set (0.00 sec) #查看相关状态呢变量 mysql> select * from information_schema.session_status where variable_name in('Last_query_cost','Handler_read_next','Innodb_pages_read','Innodb_data_reads','Innodb_pages_read'); +-------------------+----------------+ | VARIABLE_NAME | VARIABLE_VALUE | +-------------------+----------------+ | HANDLER_READ_NEXT | 11111 | #请求读的行数 | INNODB_DATA_READS | 7868409 | #数据物理读的总数 | INNODB_PAGES_READ | 7855239 | #逻辑读的总数 | LAST_QUERY_COST | 10.499000 | #SQL语句的成本COST，主要包括IO_COST和CPU_COST。 +-------------------+----------------+ 4 rows in set (0.00 sec)
#删除50w数据 mysql> delete from user limit 500000; Query OK, 500000 rows affected (3.70 sec) #分析表统计信息 mysql> analyze table user; +-----------+---------+----------+----------+ | Table | Op | Msg_type | Msg_text | +-----------+---------+----------+----------+ | test.user | analyze | status | OK | +-----------+---------+----------+----------+ 1 row in set (0.01 sec) #重置状态变量计数 mysql> flush status; Query OK, 0 rows affected (0.01 sec) mysql> select id, age ,phone from user where name like 'lyn12%'; Empty set (0.05 sec) mysql> explain select id, age ,phone from user where name like 'lyn12%'; +----+-------------+-------+-------+---------------+----------+---------+------+-------+-----------------------+ | id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra | +----+-------------+-------+-------+---------------+----------+---------+------+-------+-----------------------+ | 1 | SIMPLE | user | range | idx_name | idx_name | 82 | NULL | 22226 | Using index condition | +----+-------------+-------+-------+---------------+----------+---------+------+-------+-----------------------+ 1 row in set (0.00 sec) mysql> select * from information_schema.session_status where variable_name in('Last_query_cost','Handler_read_next','Innodb_pages_read','Innodb_data_reads','Innodb_pages_read'); +-------------------+----------------+ | VARIABLE_NAME | VARIABLE_VALUE | +-------------------+----------------+ | HANDLER_READ_NEXT | 0 | | INNODB_DATA_READS | 7868409 | | INNODB_PAGES_READ | 7855239 | | LAST_QUERY_COST | 10.499000 | +-------------------+----------------+ 4 rows in set (0.00 sec)
| 操作 | COST | 物理读次数 | 逻辑读次数 | 扫描行数 | 返回行数 | 执行时间 |
|---|---|---|---|---|---|---|
| 初始化插入100W | 10.499000 | 7868409 | 7855239 | 22226 | 11111 | 30ms |
| 100W随机删除50W | 10.499000 | 7868409 | 7855239 | 22226 | 0 | 50ms |
这也说明对普通的大表，想要通过delete数据来对表进行瘦身是不现实的，所以在任何时候不要用delete去删除数据，应该使用优雅的标记删除。
对于一个大的系统来说，需要根据业务特点去拆分子系统，每个子系统可以看做是一个service，例如美团APP，上面有很多服务，核心的服务有用户服务user-service，搜索服务search-service，商品product-service，位置服务location-service，价格服务price-service等。每个服务对应一个数据库，为该数据库创建单独账号，同时只授予DML权限且没有delete权限，同时禁止跨库访问。
#创建用户数据库并授权 create database mt_user charset utf8mb4; grant USAGE, SELECT, INSERT, UPDATE ON mt_user.* to 'w_user'@'%' identified by 't$W*g@gaHTGi123456'; flush privileges;
在MySQL数据库建模规范中有4个公共字段，基本上每个表必须有的，同时在create_time列要创建索引，有两方面的好处：
一些查询业务场景都会有一个默认的时间段，比如7天或者一个月，都是通过create_time去过滤，走索引扫描更快。
一些核心的业务表需要以T +1的方式抽取数据仓库中，比如每天晚上00:30抽取前一天的数据，都是通过create_time过滤的。
`id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键id', `is_deleted` tinyint(4) NOT NULL DEFAULT '0' COMMENT '是否逻辑删除：0：未删除，1：已删除', `create_time` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP COMMENT '创建时间', `update_time` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT '修改时间' #有了删除标记，业务接口的delete操作就可以转换为update update user set is_deleted = 1 where user_id = 1213; #查询的时候需要带上is_deleted过滤 select id, age ,phone from user where is_deleted = 0 and name like 'lyn12%';
#1. 创建归档表，一般在原表名后面添加_bak。 CREATE TABLE `ota_order_bak` ( `id` bigint(11) NOT NULL AUTO_INCREMENT COMMENT '主键', `order_id` varchar(255) DEFAULT NULL COMMENT '订单id', `ota_id` varchar(255) DEFAULT NULL COMMENT 'ota', `check_in_date` varchar(255) DEFAULT NULL COMMENT '入住日期', `check_out_date` varchar(255) DEFAULT NULL COMMENT '离店日期', `hotel_id` varchar(255) DEFAULT NULL COMMENT '酒店ID', `guest_name` varchar(255) DEFAULT NULL COMMENT '顾客', `purcharse_time` timestamp NULL DEFAULT NULL COMMENT '购买时间', `create_time` datetime DEFAULT NULL, `update_time` datetime DEFAULT NULL, `create_user` varchar(255) DEFAULT NULL, `update_user` varchar(255) DEFAULT NULL, `status` int(4) DEFAULT '1' COMMENT '状态 ： 1 正常 ， 0 删除', `hotel_name` varchar(255) DEFAULT NULL, `price` decimal(10,0) DEFAULT NULL, `remark` longtext, PRIMARY KEY (`id`), KEY `IDX_order_id` (`order_id`) USING BTREE, KEY `hotel_name` (`hotel_name`) USING BTREE, KEY `ota_id` (`ota_id`) USING BTREE, KEY `IDX_purcharse_time` (`purcharse_time`) USING BTREE, KEY `IDX_create_time` (`create_time`) USING BTREE ) ENGINE=InnoDB DEFAULT CHARSET=utf8 PARTITION BY RANGE (to_days(create_time)) ( PARTITION p201808 VALUES LESS THAN (to_days('2018-09-01')), PARTITION p201809 VALUES LESS THAN (to_days('2018-10-01')), PARTITION p201810 VALUES LESS THAN (to_days('2018-11-01')), PARTITION p201811 VALUES LESS THAN (to_days('2018-12-01')), PARTITION p201812 VALUES LESS THAN (to_days('2019-01-01')), PARTITION p201901 VALUES LESS THAN (to_days('2019-02-01')), PARTITION p201902 VALUES LESS THAN (to_days('2019-03-01')), PARTITION p201903 VALUES LESS THAN (to_days('2019-04-01')), PARTITION p201904 VALUES LESS THAN (to_days('2019-05-01')), PARTITION p201905 VALUES LESS THAN (to_days('2019-06-01')), PARTITION p201906 VALUES LESS THAN (to_days('2019-07-01')), PARTITION p201907 VALUES LESS THAN (to_days('2019-08-01')), PARTITION p201908 VALUES LESS THAN (to_days('2019-09-01')), PARTITION p201909 VALUES LESS THAN (to_days('2019-10-01')), PARTITION p201910 VALUES LESS THAN (to_days('2019-11-01')), PARTITION p201911 VALUES LESS THAN (to_days('2019-12-01')), PARTITION p201912 VALUES LESS THAN (to_days('2020-01-01'))); #2. 插入原表中无效的数据（需要跟开发同学确认数据保留范围） create table tbl_p201808 as select * from ota_order where create_time between '2018-08-01 00:00:00' and '2018-08-31 23:59:59'; #3. 跟归档表分区做分区交换 alter table ota_order_bak exchange partition p201808 with table tbl_p201808; #4. 删除原表中已经规范的数据 delete from ota_order where create_time between '2018-08-01 00:00:00' and '2018-08-31 23:59:59' limit 3000;
#1. 创建中间表 CREATE TABLE `ota_order_2020` (........) ENGINE=InnoDB DEFAULT CHARSET=utf8 PARTITION BY RANGE (to_days(create_time)) ( PARTITION p201808 VALUES LESS THAN (to_days('2018-09-01')), PARTITION p201809 VALUES LESS THAN (to_days('2018-10-01')), PARTITION p201810 VALUES LESS THAN (to_days('2018-11-01')), PARTITION p201811 VALUES LESS THAN (to_days('2018-12-01')), PARTITION p201812 VALUES LESS THAN (to_days('2019-01-01')), PARTITION p201901 VALUES LESS THAN (to_days('2019-02-01')), PARTITION p201902 VALUES LESS THAN (to_days('2019-03-01')), PARTITION p201903 VALUES LESS THAN (to_days('2019-04-01')), PARTITION p201904 VALUES LESS THAN (to_days('2019-05-01')), PARTITION p201905 VALUES LESS THAN (to_days('2019-06-01')), PARTITION p201906 VALUES LESS THAN (to_days('2019-07-01')), PARTITION p201907 VALUES LESS THAN (to_days('2019-08-01')), PARTITION p201908 VALUES LESS THAN (to_days('2019-09-01')), PARTITION p201909 VALUES LESS THAN (to_days('2019-10-01')), PARTITION p201910 VALUES LESS THAN (to_days('2019-11-01')), PARTITION p201911 VALUES LESS THAN (to_days('2019-12-01')), PARTITION p201912 VALUES LESS THAN (to_days('2020-01-01'))); #2. 插入原表中有效的数据，如果数据量在100W左右可以在业务低峰期直接插入，如果比较大，建议采用dataX来做，可以控制频率和大小，之前我这边用Go封装了dataX可以实现自动生成json文件，自定义大小去执行。 insert into ota_order_2020 select * from ota_order where create_time between '2020-08-01 00:00:00' and '2020-08-31 23:59:59'; #3. 表重命名 alter table ota_order rename to ota_order_bak; alter table ota_order_2020 rename to ota_order; #4. 插入差异数据 insert into ota_order select * from ota_order_bak a where not exists (select 1 from ota_order b where a.id = b.id); #5. ota_order_bak改造成分区表，如果表比较大不建议直接改造，可以先创建好分区表，通过dataX把导入进去即可。 #6. 后续的归档方法 #创建中间普遍表 create table ota_order_mid like ota_order; #交换原表无效数据分区到普通表 alter table ota_order exchange partition p201808 with table ota_order_mid; ##交换普通表数据到归档表的相应分区 alter table ota_order_bak exchange partition p201808 with table ota_order_mid;
这样原表和归档表都是按月的分区表，只需要创建一个中间普通表，在业务低峰期做两次分区交换，既可以删除无效数据，又能回收空，而且没有空间碎片，不会影响表上的索引及SQL的执行计划。
通过从InnoDB存储空间分布，delete对性能的影响可以看到，delete物理删除既不能释放磁盘空间，而且会产生大量的碎片，导致索引频繁分裂，影响SQL执行计划的稳定性；
同时在碎片回收时，会耗用大量的CPU，磁盘空间，影响表上正常的DML操作。
在业务代码层面，应该做逻辑标记删除，避免物理删除；为了实现数据归档需求，可以用采用MySQL分区表特性来实现，都是DDL操作，没有碎片产生。
另外一个比较好的方案采用Clickhouse，对有生命周期的数据表可以使用Clickhouse存储，利用其TTL特性实现无效数据自动清理。
相关推荐：《mysql教程》
The above is the detailed content of Why doesn't mysql use delete when deleting data?. For more information, please follow other related articles on the PHP Chinese website!