乐趣区

关于mysql:MySQL为什么错误选择代价更大的索引

欢送来到 GreatSQL 社区分享的 MySQL 技术文章,如有疑难或想学习的内容,能够在下方评论区留言,看到后会进行解答

MySQL 优化器索引抉择迷思。

高鹏(八怪)对本文亦有奉献。

1. 问题形容

群友提出问题,表里有两个列 c1、c2,别离为 INT、VARCHAR 类型,且别离创立了 unique key。

SQL 查问的条件是 WHERE c1 = ? AND c2 = ?,用 EXPLAIN 查看执行打算,发现优化器优先选择了 VARCHAR 类型的 c2 列索引。

他示意很不了解,难道不应该抉择看起来代价更小的 INT 类型的 c1 列吗?

2. 问题复现

创立测试表 t1:

[root@yejr.run]> CREATE TABLE `t1` (
  `c1` int NOT NULL AUTO_INCREMENT,
  `c2` int unsigned NOT NULL,
  `c3` varchar(20) NOT NULL,
  `c4` varchar(20) NOT NULL,
  PRIMARY KEY (`c1`),
  UNIQUE KEY `k3` (`c3`),
  UNIQUE KEY `k2` (`c2`)
) ENGINE=InnoDB;

利用 mysql_random_data_load 写入一万行数据:

mysql_random_data_load -h127.0.0.1 -uX -pX yejr t1 10000

查看执行打算:

[root@yejr.run]> EXPLAIN SELECT * FROM t1 WHERE
 c2 = 1755950419 AND c3 = 'MichaelaAnderson'\G
*************************** 1. row ***************************
           id: 1
  select_type: SIMPLE
        table: t1
   partitions: NULL
         type: const
possible_keys: k3,k2
          key: k3
      key_len: 82
          ref: const
         rows: 1
     filtered: 100.00
        Extra: NULL

能够看到优化器确实抉择了 k3 索引,而非 ” 预期 ” 的 k2 索引,这是为什么呢?

3. 问题剖析

其实起因很简略粗犷:优化器认为这两个索引抉择的代价都是一样的,只是优先选中排在后面的那个索引而已。

再建一个雷同的表 t2,只不过把 k2、k3 的索引创立程序对调下:

[root@yejr.run]> CREATE TABLE `t2` (
  `c1` int NOT NULL AUTO_INCREMENT,
  `c2` int unsigned NOT NULL,
  `c3` varchar(20) NOT NULL,
  `c4` varchar(20) NOT NULL,
  PRIMARY KEY (`c1`),
  UNIQUE KEY `k2` (`c2`),
  UNIQUE KEY `k3` (`c3`)
) ENGINE=InnoDB;

再查看执行打算:

[root@yejr.run]> EXPLAIN SELECT * FROM t2 WHERE
 c2 = 1755950419 AND c3 = 'MichaelaAnderson'\G
*************************** 1. row ***************************
           id: 1
  select_type: SIMPLE
        table: t1
   partitions: NULL
         type: const
possible_keys: k2,k3
          key: k2
      key_len: 4
          ref: const
         rows: 1
     filtered: 100.00
        Extra: NULL

咱们利用 EXPLAIN ANALYZE 来查看下两次执行打算的代价比照:

-- 查看 t1 表执行打算代价
[root@yejr.run]> EXPLAIN ANALYZE SELECT * FROM t1 WHERE
  c2 = 1755950419 AND c3 = 'MichaelaAnderson'\G
*************************** 1. row ***************************
EXPLAIN: -> Rows fetched before execution  (cost=0.00..0.00 rows=1) (actual time=0.000..0.000 rows=1 loops=1)

-- 查看 t2 表执行打算代价
[root@yejr.run]> EXPLAIN ANALYZE SELECT * FROM t2 WHERE  c2 = 1755950419 AND c3 = 'MichaelaAnderson'\G
*************************** 1. row ***************************
EXPLAIN: -> Rows fetched before execution  (cost=0.00..0.00 rows=1) (actual time=0.000..0.000 rows=1 loops=1)

能够看到,很显著代价都是一样的。

再利用 OPTIMIZE_TRACE 查看执行打算,也能看到两个 SQL 的代价是一样的:

...
          {
            "rows_estimation": [
              {
                "table": "`t1`",
                "rows": 1,
                "cost": 1,
                "table_type": "const",
                "empty": false
              }
            ]
          },
...

所以,优化器认为抉择哪个索引都是一样的,就看哪个索引排序更靠前。

从执行 SELECT 时的 debug trace 后果也能佐证:

-- 1、T1 表,k3 索引在后面
  PRIMARY KEY (`c1`),
  UNIQUE KEY `k3` (`c3`),
  UNIQUE KEY `k2` (`c2`)
  
T@2: | | | | | | | | opt: (null): starting struct
T@2: | | | | | | | | opt: table: "`t1`"
T@2: | | | | | | | | opt: field: "c3"   (C3 在后面,因而最初应用 k3)
T@2: | | | | | | | | >convert_string
T@2: | | | | | | | | | >alloc_root
T@2: | | | | | | | | | | enter: root: 0x40a8068
T@2: | | | | | | | | | | exit: ptr: 0x4b41ab0
T@2: | | | | | | | | | <alloc_root 304
T@2: | | | | | | | | <convert_string 2610
T@2: | | | | | | | | opt: equals: "'Louise Garrett'" 
T@2: | | | | | | | | opt: null_rejecting: 0
T@2: | | | | | | | | opt: (null): ending struct
T@2: | | | | | | | | opt: Key_use: optimize= 0 used_tables=0x0 ref_table_rows= 18446744073709551615 keypart_map= 1
T@2: | | | | | | | | opt: (null): starting struct
T@2: | | | | | | | | opt: table: "`t1`"
T@2: | | | | | | | | opt: field: "c2"
T@2: | | | | | | | | opt: equals: "22896242"
T@2: | | | | | | | | opt: null_rejecting: 0
T@2: | | | | | | | | opt: null_rejecting: 0
T@2: | | | | | | | | opt: (null): ending struct
T@2: | | | | | | | | opt: Key_use: optimize= 0 used_tables=0x0 ref_table_rows= 18446744073709551615 keypart_map= 1
T@2: | | | | | | | | opt: (null): starting struct
T@2: | | | | | | | | opt: table: "`t1`"
T@2: | | | | | | | | opt: field: "c2"
T@2: | | | | | | | | opt: equals: "22896242"
T@2: | | | | | | | | opt: null_rejecting: 0
T@2: | | | | | | | | opt: (null): ending struct
T@2: | | | | | | | | opt: ref_optimizer_key_uses: ending struct
T@2: | | | | | | | | opt: (null): ending struct

-- 2、T2 表,k2 索引在后面
  PRIMARY KEY (`c1`),
  UNIQUE KEY `k2` (`c2`),
  UNIQUE KEY `k3` (`c3`)
  
T@2: | | | | | | | | opt: (null): starting struct
T@2: | | | | | | | | opt: table: "`t2`"
T@2: | | | | | | | | opt: field: "c2"(C2 在后面因而应用 k2 索引)T@2: | | | | | | | | opt: equals: "22896242"
T@2: | | | | | | | | opt: null_rejecting: 0
T@2: | | | | | | | | opt: (null): ending struct
T@2: | | | | | | | | opt: Key_use: optimize= 0 used_tables=0x0 ref_table_rows= 18446744073709551615 keypart_map= 1
T@2: | | | | | | | | opt: (null): starting struct
T@2: | | | | | | | | opt: table: "`t2`"
T@2: | | | | | | | | opt: field: "c3"
T@2: | | | | | | | | >convert_string
T@2: | | | | | | | | | >alloc_root
T@2: | | | | | | | | | | enter: root: 0x40a8068
T@2: | | | | | | | | | | exit: ptr: 0x4b41ab0
T@2: | | | | | | | | | <alloc_root 304
T@2: | | | | | | | | <convert_string 2610
T@2: | | | | | | | | opt: equals: "'Louise Garrett'"
T@2: | | | | | | | | opt: null_rejecting: 0
T@2: | | | | | | | | opt: (null): ending struct
T@2: | | | | | | | | opt: ref_optimizer_key_uses: ending struct
T@2: | | | | | | | | opt: (null): ending struct

4. 问题延长

到这里,咱们不禁有疑难,这两个索引的代价真的是一样吗?

就让咱们用 mysqlslap 来做个简略比照测试吧:

-- 测试 1:对 c2 列随机 point select
mysqlslap -hlocalhost -uroot -Smysql.sock --no-drop --create-schema X -i 3 --number-of-queries 1000000 -q "set @xid = cast(round(rand()*2147265929) as unsigned); select * from t1 where c2 = @xid" -c 8
...
    Average number of seconds to run all queries: 9.483 seconds
...


-- 测试 2:对 c3 列随机 point select
mysqlslap -hlocalhost -uroot -Smysql.sock --no-drop --create-schema X -i 3 --number-of-queries 1000000 -q "set @xid = concat('u',cast(round(rand()*2147265929) as unsigned)); select * from t1 where c3 = @xid" -c 8
...
    Average number of seconds to run all queries: 10.360 seconds
...

能够看到,如果是走 c3 列索引,耗时会比走 c2 列索引多进去约 7% ~ 9%(在我的环境下测试的后果,不同环境、不同数据量可能也不同)。

看来,MySQL 优化器还是有必要进一步提高的哟:)

测试应用版本:GreatSQL 8.0.25(MySQL 5.6.39 后果亦是如此)。

Enjoy GreatSQL :)

本文由博客一文多发平台 OpenWrite 公布!

退出移动版