测试环境：win10系统+16G内存+mysql5.7.25

1.创建测试表(表结构如下)

2.插入数据方法

@Override

@Transactional(rollbackFor = Exception.class)

public long batchInsert() {

//初始化1000000条数据

List> data = createList(1000000);

Long startTime = System.currentTimeMillis();

String addSql = "insert into aaa (first, second, three, four,five) VALUES(?,?,?,?,?)";

String updateSql = "update aaa set first = ?,second=?,three=?,four=?,five=? where id = ?";

jdbcTemplate.batchUpdate(addSql, new BatchPreparedStatementSetter() {

@Override

public void setValues(PreparedStatement ps, int i) throws SQLException {

ps.setString(1, data.get(i).get("first"));

ps.setString(2, data.get(i).get("second"));

ps.setString(3, data.get(i).get("three"));

ps.setString(4, data.get(i).get("four"));

ps.setString(5, data.get(i).get("five"));

}

@Override

public int getBatchSize() {

return data.size();

}

});

Long endTime = System.currentTimeMillis();

System.out.println("插入"+data.size()+"条数据用时：" + (endTime - startTime) + "毫秒");

System.out.println("插入"+data.size()+"条数据用时：" + (endTime - startTime) / 1000 + "秒");

return data.size();

}

3.执行结果：

4.更新数据方法

@Override

@Transactional(rollbackFor = Exception.class)

public long batchInsert() {

List> data = createList(1000000);

Long startTime = System.currentTimeMillis();

String updateSql = "update aaa set first = ?,second=?,three=?,four=?,five=? where id = ?";

jdbcTemplate.batchUpdate(updateSql, new BatchPreparedStatementSetter() {

@Override

public void setValues(PreparedStatement ps, int i) throws SQLException {

ps.setString(1, data.get(i).get("first"));

ps.setString(2, data.get(i).get("second"));

ps.setString(3, data.get(i).get("three"));

ps.setString(4, data.get(i).get("four"));

ps.setString(5, data.get(i).get("five"));

ps.setInt(6, 3212021 + i);

}

@Override

public int getBatchSize() {

return data.size();

}

});

Long endTime = System.currentTimeMillis();

System.out.println("更新"+data.size()+"条数据用时：" + (endTime - startTime) + "毫秒");

System.out.println("更新"+data.size()+"条数据用时：" + (endTime - startTime) / 1000 + "秒");

return data.size();

}

5.更新结果，一百万条数据五个字段全部被更新

6.注意

要开启批处理，方法一定要开启事务，数据量再大的话可多线程处理