准备:
1.表结构
CREATE TABLE `t_user` (
`id` varchar(32) CHARACTER SET utf8 NOT NULL COMMENT '主键',
`name` varchar(50) CHARACTER SET utf8 DEFAULT NULL COMMENT '用户名',
`del_flag` char(1) CHARACTER SET utf8 DEFAULT NULL COMMENT '删除标示',
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1;
2.1 jdbc.properties配置
mysql.driver=com.mysql.jdbc.Driver
mysql.url=jdbc:mysql://127.0.0.1:3306/ssm
mysql.username=root
mysql.password=admin
#定义初始连接数
mysql.initialSize=1
#定义最大连接数
mysql.maxActive=20
#定义最大空闲
mysql.maxIdle=20
#定义最小空闲
mysql.minIdle=1
#定义最长等待时间
mysql.maxWait=60000
2.2 spring-mybatis.xml配置
<context:component-scan base-package="com.win.ssm"/>
<context:property-placeholder location="classpath:jdbc.properties"/>
<bean id="dataSource" class="org.apache.commons.dbcp.BasicDataSource" destroy-method="close">
<property name="driverClassName" value="${mysql.driver}"/>
<property name="url" value="${mysql.url}"/>
<property name="username" value="${mysql.username}"/>
<property name="password" value="${mysql.password}"/>
<!-- 初始化链接大小-->
<property name="initialSize" value="${mysql.initialSize}"/>
<!-- 连接池最大数量-->
<property name="maxActive" value="${mysql.maxActive}"/>
<!-- 连接池最大空闲-->
<property name="maxIdle" value="${mysql.maxIdle}"/>
<!-- 连接池最小空闲 -->
<property name="minIdle" value="${mysql.minIdle}"></property>
<!-- 获取连接最大等待时间-->
<property name="maxWait" value="${mysql.maxWait}"/>
</bean>
<!-- spring与mybatis整合类 -->
<bean id="sqlSessionFactory" class="org.mybatis.spring.SqlSessionFactoryBean">
<property name="dataSource" ref="dataSource"/>
<!-- 查找接口的别名 -->
<property name="typeAliasesPackage" value="com.win"/>
<!-- 自动扫描mapping.xml文件-->
<property name="mapperLocations" value="classpath:/mapping/*.xml"/>
</bean>
<bean id="sqlSessionTemplate" class="org.mybatis.spring.SqlSessionTemplate">
<constructor-arg index="0" ref="sqlSessionFactory" />
<!--<constructor-arg index="1" value="BATCH" />-->
</bean>
<!-- 扫描DAO接口 -->
<bean id="mapperScannerConfigurer" class="org.mybatis.spring.mapper.MapperScannerConfigurer">
<property name="basePackage" value="com.win.ssm.dao"/>
<property name="sqlSessionFactoryBeanName" value="sqlSessionFactory"/>
</bean>
<!-- 事务管理 -->
<bean class="org.springframework.jdbc.datasource.DataSourceTransactionManager">
<property name="dataSource" ref="dataSource"/>
</bean>
第一种:普通for循环插入
①junit类
@Test
public void testInsertBatch2() throws Exception {
long start = System.currentTimeMillis();
User user;
SqlSession sqlSession = sqlSessionTemplate.getSqlSessionFactory().openSession(false);
UserDao mapper = sqlSession.getMapper(UserDao.class);
for (int i = 0; i < 500; i++) {
user = new User();
user.setId("test" + i);
user.setName("name" + i);
user.setDelFlag("0");
mapper.insert(user);
}
sqlSession.commit();
long end = System.currentTimeMillis();
System.out.println("---------------" + (start - end) + "---------------");
}
②xml配置
<insert id="insert">
INSERT INTO t_user (id, name, del_flag)
VALUES(#{id}, #{name}, #{delFlag})
</insert>
第二种:mybatis BATCH模式插入
①junit类
@Test
public void testInsertBatch2() throws Exception {
long start = System.currentTimeMillis();
User user;
SqlSession sqlSession = sqlSessionTemplate.getSqlSessionFactory().openSession(ExecutorType.BATCH, false);//跟上述sql区别
UserDao mapper = sqlSession.getMapper(UserDao.class);
for (int i = 0; i < 500; i++) {
user = new User();
user.setId("test" + i);
user.setName("name" + i);
user.setDelFlag("0");
mapper.insert(user);
}
sqlSession.commit();
long end = System.currentTimeMillis();
System.out.println("---------------" + (start - end) + "---------------");
}
②xml配置与第一种②中使用相同
第三种:foreach方式插入
①junit类
@Test
public void testInsertBatch() throws Exception {
long start = System.currentTimeMillis();
List<User> list = new ArrayList<>();
User user;
for (int i = 0; i < 10000; i++) {
user = new User();
user.setId("test" + i);
user.setName("name" + i);
user.setDelFlag("0");
list.add(user);
}
userService.insertBatch(list);
long end = System.currentTimeMillis();
System.out.println("---------------" + (start - end) + "---------------");
}
②xml配置
<insert id="insertBatch">
INSERT INTO t_user
(id, name, del_flag)
VALUES
<foreach collection ="list" item="user" separator =",">
(#{user.id}, #{user.name}, #{user.delFlag})
</foreach >
</insert>
特别注意:mysql默认接受sql的大小是1048576(1M),即第三种方式若数据量超过1M会报如下异常:(可通过调整MySQL安装目录下的my.ini文件中[mysqld]段的"max_allowed_packet = 1M")
nested exception is com.mysql.jdbc.PacketTooBigException: Packet for query is too large (5677854 > 1048576).
You can change this value on the server by setting the max_allowed_packet' variable.
结果对比:
第一种 | 第二种 | 第三种 | |
500条 | 7742 | 7388 | 622 |
1000条 | 15290 | 15078 | 746 |
5000条 | 78011 | 177350 | 1172 |
10000条 | 397472 | 201180 | 1205 |
时间有限测试数据较少,有兴趣可以自己测试以下。(不清楚为什么BATCH有时候比单条循环插入还耗时间,请知道的大神不吝赐教,感谢!)