| 
                            
                                  1 数据准备1.1 新建数据表
	
		
			| 1 2 3 4 5 6 7 8 9 10 11 12 13 | CREATE TABLE `player` (   `id` bigint(20) NOT NULL AUTO_INCREMENT COMMENT '主键',   `player_id` varchar(256) NOT NULL COMMENT '运动员编号',   `player_name` varchar(256) NOT NULL COMMENT '运动员名称',   `height` int(11) NOT NULL COMMENT '身高',   `weight` int(11) NOT NULL COMMENT '体重',   `type` varchar(256) DEFAULT '0' COMMENT '球员类型',   `game_performance` text COMMENT '最近一场比赛表现',   PRIMARY KEY (`id`),   KEY `idx_name_height_weight` (`player_name`,`height`,`weight`),   KEY `idx_type` (`type`),   KEY `idx_height` (`height`) ) ENGINE=InnoDB AUTO_INCREMENT=1 DEFAULT CHARSET=utf8 |  以上数据表声明三个索引: 
	联合索引:idx_name_height_weight普通索引:idx_type普通索引:idx_height 1.2 新增100万条数据
	
		
			| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 | @SpringBootTest(classes = TestApplication.class) @RunWith(SpringJUnit4ClassRunner.class) public class PlayerServiceTest {       @Resource     private PlayerRepository playerRepository;       @Test     public void initBigData() {         for (int i = 0; i < 1000000; i++) {             PlayerEntity entity = new PlayerEntity();             entity.setPlayerId(UUID.randomUUID().toString());             entity.setPlayerName("球员_" + System.currentTimeMillis());             entity.setType("0");             entity.setWeight(150);             entity.setHeight(188);             entity.setGamePerformance("{\"runDistance\":8900.0,\"passSuccess\":80.12,\"scoreNum\":3}");             playerRepository.insert(entity);         }     } } |  2 基础知识2.1 explain type执行计划中访问类型是重要分析指标: 
 2.2 explain ExtraExtra表示执行计划扩展信息: 
 3 索引失效场景本章节介绍索引失效十种场景: 
	查询类型错误索引列参与运算错误使用通配符未用到覆盖索引OR连接无索引字段MySQL放弃使用索引联合索引失效
	
	 3.1 查询类型错误3.1.1 失效场景
	
		
			| 1 | explain select * from player where type = 0 |  
 3.1.2 解决方案数据表定义type字段为varchar类型,查询必须使用相同类型: 
 3.2 索引列参与运算3.2.1 失效场景
	
		
			| 1 | explain select * from player where height + 1 > 189 |  
 3.2.2 解决方案
	
		
			| 1 | explain select * from player where height > 188 |  
 3.3 MySQL放弃使用索引3.3.1 失效场景MySQL发现如果使用索引性能低于全表扫描则放弃使用索引。例如在表中100万条数据height字段值全部是188,所以执行如下语句时放弃使用索引: 
	
		
			| 1 | explain select * from player where height > 187 |  
 3.3.2 解决方案一调整查询条件值: 
	
		
			| 1 | explain select * from player where height > 188 |  
 3.3.3 解决方案二强制指定索引,这种方法不一定可以提升性能: 
 3.4 错误使用通配符3.4.1 数据准备避免出现3.3章节失效问题此处修改一条数据: 
	
		
			| 1 | update player set player_name = '测试球员' where id = 1 |  
 3.4.2 失效场景一
	
		
			| 1 | explain select * from player where player_name like '%测试' |  
 3.4.3 失效场景二
	
		
			| 1 | explain select * from player where player_name like '%测试%' |  
 3.4.4 解决方案
	
		
			| 1 | explain select * from player where player_name like '测试%' |  
 3.5 OR连接无索引字段3.5.1 失效场景type有索引,weight无索引: 
	
		
			| 1 | explain select * from player where type = '0' or weight = 150 |  
 3.5.2 解决方案weight新增索引,union拼装查询数据 
	
		
			| 1 2 3 4 | explain select * from player where type = '0' union select * from player where weight = 150 |  
 3.6 未用到覆盖索引3.6.1 失效场景Using index condition表示使用索引,但是需要回表查询 
	
		
			| 1 | explain select * from player where player_name like '测试%' |  
 3.6.2 解决方案覆盖索引含义是查询时索引列完全包含查询列,查询过程无须回表(需要在同一棵索引树)性能得到提升。Using Index; Using where表示使用覆盖索引并且用where过滤查询结果: 
	
		
			| 1 | explain select id,player_name,height,weight from player where player_name like '测试%' |  
 3.7 联合索引失效3.7.1 完整使用联合索引idx_name_height_weight完整使用key_len=778: 
	
		
			| 1 | explain select * from player where player_name = '球员_1682577684751' and height = 188 and weight = 150 |  
 3.7.2 失效场景一:索引不完整weight不在查询条件,所以只用到idx_name_height,所以key_len= 774: 
	
		
			| 1 | explain select * from player where player_name = '球员_1682577684751' and height = 188 |  
 3.7.3 失效场景二:索引中断height不在查询条件,所以只用到idx_name,所以key_len= 770: 
	
		
			| 1 | explain select * from player where player_name = '球员_1682577684751' and weight = 150 |  
 3.7.4 失效场景三:非等值匹配height非等值匹配,所以只用到idx_name_height,所以key_length=774: 
	
		
			| 1 | explain select * from player where player_name='球员_1682577684751' and height > 188 and weight = 150 |  
 3.7.5 失效场景四:最左索引缺失player_name最左索引不在查询条件,全表扫描 
	
		
			| 1 | explain select * from player where weight = 150 |  
 4 文章总结本文第一进行测试数据准备,第二介绍执行计划相关知识,第三介绍索引失效10种场景:查询类型错误,索引列参与运算,错误使用通配符,未用到覆盖索引,OR连接无索引字段,MySQL放弃使用索引,联合索引中索引不完整,索引中断,非等值匹配,最左索引缺失。 
 |