文章目錄

• 生猛干貨
• DB Version
• Table
• 數據量
• 案例一 ：explain select * from employees where name = 'LiLei' and position = 'dev' order by age
• 案例二： explain select * from employees where name = 'LiLei' order by position
• 案例三：explain select * from employees where name = 'LiLei' order by age , position
• 案例四：explain select * from employees where name = 'LiLei' order by position , age
• 案例五：explain select * from employees where name = 'LiLei' and age = 18 order by position , age ;
• 案例六：explain select * from employees where name = 'LiLei' order by age asc , position desc ;
• 案例七：explain select * from employees where name in ('HanMeiMei' , 'LiLei') order by age , position ;
• 案例八： explain select * from employees where name > 'HanMeiMei' order by name ;
• group by 優化
• 小結
• 搞定MySQL

---

生猛干貨

帶你搞定MySQL實戰，輕松對應海量業務處理及高并發需求，從容應對大場面試

---

DB Version

mysql> select version(); +-----------+ | version() | +-----------+ | 5.7.28 | +-----------+ 1 row in setmysql>

---

Table

CREATE TABLE `employees` (`id` int(11) NOT NULL AUTO_INCREMENT,`name` varchar(24) NOT NULL DEFAULT '' COMMENT '姓名',`age` int(11) NOT NULL DEFAULT '0' COMMENT '年齡',`position` varchar(20) NOT NULL DEFAULT '' COMMENT '職位',`hire_time` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP COMMENT '入職時間',PRIMARY KEY (`id`),KEY `idx_name_age_position` (`name`,`age`,`position`) USING BTREE ) ENGINE=InnoDB AUTO_INCREMENT=1 DEFAULT CHARSET=utf8 COMMENT='員工記錄表';

兩個索引

主鍵索引

二級索引 KEY idx_name_age_position (name,age,position) USING BTREE

重點就是這個二級索引 ，記號了哈。

---

數據量

mysql> select count(1) from employees ; +----------+ | count(1) | +----------+ | 100002 | +----------+ 1 row in setmysql>

---

案例一 ：explain select * from employees where name = ‘LiLei’ and position = ‘dev’ order by age

explain select * from employees where name = 'LiLei' and position = 'dev' order by age ;

先想一下這個order by 會不會走索引 ？

會走索引

原因呢 ？

腦海中要有這個聯合索引在MySQL底層的B+Tree的數據結構 ， 索引 排好序的數據結構。

name = ‘LiLei’ and position = ‘dev’ order by age

name 為 LiLei , name 確定的情況下， age 肯定是有序的 ，age 有序不能保證position 有序

所以 這個order by age 是可以走索引的

繼續分析下這個explain

mysql> explain select * from employees where name = 'LiLei' and position = 'dev' order by age ; +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+-----------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+-----------------------+ | 1 | SIMPLE | employees | NULL | ref | idx_name_age_position | idx_name_age_position | 74 | const | 1 | 10 | Using index condition | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+-----------------------+ 1 row in set

order by 走的索引 是不會體現在key_len上的, 這個74 = 3 * 24 + 2 ， 是計算的name 。 最左匹配原則 ，中間字段不能斷，因此查詢用到了name索引。

但是 Extra直接里面可以看出來 Using index condition ，說明age索引列用在了排序過程中 。 如果沒有走索引的話，那就是 Using FileSort 了

接下來繼續看幾個例子，加深理解，重點是腦海中的 索引B+Tree結構

---

案例二： explain select * from employees where name = ‘LiLei’ order by position

mysql> explain select * from employees where name = 'LiLei' order by position ;

想一想，這個order by 會走索引嗎？

我們來看下索引 KEY idx_name_age_position (name,age,position) USING BTREE

再來看下查詢SQL

where name = 'LiLei' order by position ;

name = LiLei , name 值能確定下來， 符合最左匹配原則 所以查詢會走索引 ， 用了聯合索引中的name字段， key len = 74 . 所以 Using index condition

order by position , 在索引中 中間缺失了age , 用position ，跳過了age , 那索引樹能是有序的嗎？ 肯定不是。。。所以 position肯定不是排好序的 ， 無法走索引排序，因此 Extra信息 有 Using filesort

來看下執行計劃

mysql> explain select * from employees where name = 'LiLei' order by position ; +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ | 1 | SIMPLE | employees | NULL | ref | idx_name_age_position | idx_name_age_position | 74 | const | 1 | 100 | Using index condition; Using filesort | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ 1 row in setmysql>

正如分析~

有感覺了嗎？ 再來看一個

---

案例三：explain select * from employees where name = ‘LiLei’ order by age , position

這個SQL和案例二的很相似 ， 僅僅在排序的時候在前面多了一個age字段參與排序 ， 那分析分析 order by 會走索引嗎

mysql> explain select * from employees where name = 'LiLei' order by age , position ;

時刻不要那個索引樹 ，來分析一下

name = LiLei , name 固定，結合 建立的索引， 最左原則，所以查詢肯定會走聯合索引中的部分索引 name .

在name都是LiLei 的情況下 ， order by age , position 結合索引樹 ，age和position用于排序 也是有序的，應該不會走using filesort

我們來看下執行計劃

mysql> explain select * from employees where name = 'LiLei' order by age , position ; +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+-----------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+-----------------------+ | 1 | SIMPLE | employees | NULL | ref | idx_name_age_position | idx_name_age_position | 74 | const | 1 | 100 | Using index condition | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+-----------------------+ 1 row in setmysql>

---

案例四：explain select * from employees where name = ‘LiLei’ order by position , age

再分析一個，和案例上也很像。 把 order by的排序順序 調整一下，我們來分析一下 order by會不會走索引

explain select * from employees where name = 'LiLei' order by position , age ;

mysql> explain select * from employees where name = 'LiLei' order by position , age ; +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ | 1 | SIMPLE | employees | NULL | ref | idx_name_age_position | idx_name_age_position | 74 | const | 1 | 100 | Using index condition; Using filesort | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ 1 row in setmysql>

咦 ， 執行計劃中有 using filesort

為什么呢？

看看我們二級索引的建立的字段順序 ， 創建順序為name,age,position，但是排序的時候age和position顛倒位置了， 那排好序的特性肯定就無法滿足了，那你讓MySQL怎么走索引？

---

案例五：explain select * from employees where name = ‘LiLei’ and age = 18 order by position , age ;

這個order by 會走索引嗎？

mysql> explain select * from employees where name = 'LiLei' and age = 18 order by position , age ; +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------------+------+----------+-----------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------------+------+----------+-----------------------+ | 1 | SIMPLE | employees | NULL | ref | idx_name_age_position | idx_name_age_position | 78 | const,const | 1 | 100 | Using index condition | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------------+------+----------+-----------------------+ 1 row in setmysql>

走了dx_name_age_position 索引中的 name 和 age , order by 其實也走了索引，你看extra中并沒有 using filesort ,因為age為常量，在排序中被MySQL優化了，所以索引未顛倒，不會出現Using filesort

---

案例六：explain select * from employees where name = ‘LiLei’ order by age asc , position desc ;

mysql> explain select * from employees where name = 'LiLei' order by age asc , position desc ; +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ | 1 | SIMPLE | employees | NULL | ref | idx_name_age_position | idx_name_age_position | 74 | const | 1 | 100 | Using index condition; Using filesort | +----+-------------+-----------+------------+------+-----------------------+-----------------------+---------+-------+------+----------+---------------------------------------+ 1 row in set

我們可以看到雖然排序的字段列與建立索引的順序一樣， order by默認升序排列，而SQL中的 position desc變成了降序排列，導致與索引的排序方式不同，從而產生Using filesort。

Note: Mysql8以上版本有降序索引可以支持該種查詢方式。

---

案例七：explain select * from employees where name in (‘HanMeiMei’ , ‘LiLei’) order by age , position ;

mysql> explain select * from employees where name in ('HanMeiMei' , 'LiLei') order by age , position ; +----+-------------+-----------+------------+-------+-----------------------+-----------------------+---------+------+------+----------+---------------------------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+-------+-----------------------+-----------------------+---------+------+------+----------+---------------------------------------+ | 1 | SIMPLE | employees | NULL | range | idx_name_age_position | idx_name_age_position | 74 | NULL | 2 | 100 | Using index condition; Using filesort | +----+-------------+-----------+------------+-------+-----------------------+-----------------------+---------+------+------+----------+---------------------------------------+ 1 row in setmysql>

對order by 來講 ，多個相等的條件也是 范圍查詢。 既然是范圍查詢， 可能對于每個值在索引中是有序的，但多個合并在一起，就不是有序的了，所以 using filesort .

---

案例八： explain select * from employees where name > ‘HanMeiMei’ order by name ;

mysql> explain select * from employees where name > 'HanMeiMei' order by name ; +----+-------------+-----------+------------+------+-----------------------+------+---------+------+-------+----------+-----------------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+------+-----------------------+------+---------+------+-------+----------+-----------------------------+ | 1 | SIMPLE | employees | NULL | ALL | idx_name_age_position | NULL | NULL | NULL | 96845 | 50 | Using where; Using filesort | +----+-------------+-----------+------------+------+-----------------------+------+---------+------+-------+----------+-----------------------------+ 1 row in setmysql>

MySQL自己內部有一套優化機制，且數據量不同、版本不一樣，結果也可能有差異

一般情況下， 聯合索引第一個字段用范圍不一定會走索引 ， 可以采用 覆蓋索引進行優化，避免回表帶來的性能開銷 。

mysql> explain select namefrom employees where name > 'a' order by name ; +----+-------------+-----------+------------+-------+-----------------------+-----------------------+---------+------+-------+----------+--------------------------+ | id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra | +----+-------------+-----------+------------+-------+-----------------------+-----------------------+---------+------+-------+----------+--------------------------+ | 1 | SIMPLE | employees | NULL | range | idx_name_age_position | idx_name_age_position | 74 | NULL | 48422 | 100 | Using where; Using index | +----+-------------+-----------+------------+-------+-----------------------+-----------------------+---------+------+-------+----------+--------------------------+ 1 row in setmysql>

---

group by 優化

• group by與order by類似，其實質是先排序后分組，遵照索引創建順序的最左前綴法則。

• 對于group by的優化如果不需要排序的可以加上order by null禁止排序。

• where高于having，能寫在where中的限定條件就不要去having限定了。

---

小結

• MySQL支持兩種方式的排序filesort和index，Using index是指MySQL掃描索引本身完成排序

• order by滿足兩種情況會使用Using index
  A: order by語句使用索引最左前列。
  B: 使用where子句與order by子句條件列組合滿足索引最左前列

• 盡量在索引列上完成排序，遵循索引建立（索引創建的順序）時的最左前綴法則

• 如果order by的條件不在索引列上，就會產生Using filesort

• 能用覆蓋索引盡量用覆蓋索引

---

搞定MySQL

總結

以上是生活随笔為你收集整理的MySQL - order by和 group by 优化初探的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。