2019独角兽企业重金招聘Python工程师标准>>> 
MySQL/InnoDB的并发插入Concurrent Insert
表people建表语句：ENGINE=InnoDB
CREATE TABLE people (
person_id BIGINT NOT NULL AUTO_INCREMENT,
first_name VARCHAR(20),
last_name VARCHAR(20),
PRIMARY KEY (person_id)
);
两个会话：会话一和会话二
在会话一中，执行如下sql语句：
mysql> use local_database;
Database changed
mysql> select @@autocommit;
+--------------+
| @@autocommit |
+--------------+
| 1 |
+--------------+
1 row in set (0.00 sec)
mysql> set autocommit = 0;
Query OK, 0 rows affected (0.00 sec)
mysql> select @@autocommit;
+--------------+
| @@autocommit |
+--------------+
| 0 |
+--------------+
1 row in set (0.00 sec)
mysql> select * from people;
+-----------+------------+-----------+
| person_id | first_name | last_name |
+-----------+------------+-----------+
| 1 | 1111 | 1111 |
+-----------+------------+-----------+
1 row in set (0.00 sec)
mysql> begin;
Query OK, 0 rows affected (0.00 sec)
mysql> insert into people (first_name,last_name) values ('1111','1111');
Query OK, 1 row affected (0.12 sec)
mysql> select * from people;
+-----------+------------+-----------+
| person_id | first_name | last_name |
+-----------+------------+-----------+
| 1 | 1111 | 1111 |
| 3 | 1111 | 1111 |
+-----------+------------+-----------+
2 rows in set (0.00 sec)
上面这些sql 语句最终的操作就是手动开启了一个事务，然后提交了一个insert语句，注意没有手动提交事务。。
此时在另一个会话二中进行如下操作
mysql> use local_database;
Database changed
mysql> select @@autocommit;
+--------------+
| @@autocommit |
+--------------+
| 1 |
+--------------+
1 row in set (0.00 sec)
mysql> set autocommit = 0;
Query OK, 0 rows affected (0.00 sec)
mysql> select @@autocommit;
+--------------+
| @@autocommit |
+--------------+
| 0 |
+--------------+
1 row in set (0.00 sec)
mysql> select * from people;
+-----------+------------+-----------+
| person_id | first_name | last_name |
+-----------+------------+-----------+
| 1 | 1111 | 1111 |
+-----------+------------+-----------+
1 row in set (0.00 sec)
mysql> begin;
Query OK, 0 rows affected (0.00 sec)
mysql> insert into people (first_name,last_name) values ('2222','2222');
Query OK, 1 row affected (0.00 sec)
mysql> select * from people;
+-----------+------------+-----------+
| person_id | first_name | last_name |
+-----------+------------+-----------+
| 1 | 1111 | 1111 |
| 4 | 2222 | 2222 |
+-----------+------------+-----------+
2 rows in set (0.00 sec)
上面的sql语句最终的操作是手动开启了一个事务，执行了insert语句，而没有提交事务，然后select查看只会看到当前会话的操作结果，而没有会话一的操作结果。这就是mysql默认事务隔离级别——可重复读。
做到这里，我验证的不是可重复读的事务隔离级别，我其实想验证的是会话在事务内执行insert语句会不会给表加锁（会给表加锁，AUTO-INC lock），通过结果显示，可知，不同的会话在事务内执行insert语句，而不会阻塞其他会话事务内的insert语句。
但最终我产生了疑问，MySQL/InnoDB是如何处理并发插入的。
MySQL/InnoDB的并发插入
说起auto_increment的并发插入，就要理解auto_increment的机制了。见文章：
http://my.oschina.net/xinxingegeya/blog/341991
http://my.oschina.net/xinxingegeya/blog/342075
下面我写了两个程序测试并发插入，一个是单线程的写入10w条数据，一个是100条线程单个线程写入1000条数据。可以运行一下程序做一个对比。
建表语句
CREATE TABLE people_thread (
person_id BIGINT NOT NULL AUTO_INCREMENT,
first_name VARCHAR(20),
last_name VARCHAR(20),
thread_name VARCHAR(20),
PRIMARY KEY (person_id)
);总之还是多线程的效率比较高：测试数据如下
单线程的运行时间为：-- 16656ms
多线程的最长的运行时间为：-- 12632ms
JDBCTest2.java
package com.lyx.other;
import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.PreparedStatement;
import java.sql.SQLException;
public class JDBCTest2 {
public static void main(String args[]) throws SQLException {
long startTime = System.currentTimeMillis(); // 获取开始时间
Connection conn = null;
PreparedStatement ps = null;
String sql = "insert into people_thread (first_name ,last_name ,"
+ "thread_name) values (?,?,?)";
try {
conn = getConnection();
conn.setAutoCommit(false);
ps = conn.prepareStatement(sql);
for (int i = 0; i < 100000; i++) {
ps.setString(1, Integer.toString(i));
ps.setString(2, Integer.toString(i));
ps.setString(3, Integer.toString(i));
ps.addBatch();
// 一批提交一次
if (i % 10 == 0) {
ps.executeBatch();
ps.clearBatch();
}
}
ps.executeBatch();
conn.commit();
} catch (Exception e) {
// TODO: handle exception
e.printStackTrace();
} finally {
if (conn != null) {
try {
conn.close();
} catch (SQLException e) {
e.printStackTrace();
}
}
}
long endTime = System.currentTimeMillis(); // 获取结束时间
System.out.println("程序运行时间： " + (endTime - startTime) + "ms");
}
/* 获取数据库连接的函数 */
public static Connection getConnection() {
Connection con = null; // 创建用于连接数据库的Connection对象
try {
Class.forName("com.mysql.jdbc.Driver");// 加载Mysql数据驱动
con = DriverManager.getConnection(
"jdbc:mysql://localhost:3306/local_database", "root",
"034039");// 创建数据连接
} catch (Exception e) {
System.out.println("数据库连接失败" + e.getMessage());
}
return con; // 返回所建立的数据库连接
}
}JDBCTest3.java
package com.lyx.other;
import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.PreparedStatement;
import java.sql.SQLException;
public class JDBCTest3 {
public static void main(String[] args) {
// 创建10个线程执行插入操作
for (int i = 0; i < 100; i++) {
final int n = i;
Runnable task = new Runnable() {
public void run() {
Thread.currentThread().setName("thread_" + n);
insertByBatch(Thread.currentThread().getName());
}
};
Thread thread = new Thread(task);
thread.start();
}
}
public static void insertByBatch(String name) {
long startTime = System.currentTimeMillis(); // 获取开始时间
Connection conn = null;
PreparedStatement ps = null;
String sql = "insert into people_thread (first_name ,last_name ,"
+ "thread_name) values (?,?,?)";
try {
conn = getConnection();
conn.setAutoCommit(false);
ps = conn.prepareStatement(sql);
for (int i = 0; i < 1000; i++) {
ps.setString(1, Integer.toString(i));
ps.setString(2, Integer.toString(i));
ps.setString(3, name);
ps.addBatch();
// 一批提交一次
if (i % 10 == 0) {
ps.executeBatch();
ps.clearBatch();
}
}
ps.executeBatch();
conn.commit();
} catch (Exception e) {
// TODO: handle exception
e.printStackTrace();
} finally {
if (conn != null) {
try {
conn.close();
} catch (SQLException e) {
e.printStackTrace();
}
}
}
long endTime = System.currentTimeMillis(); // 获取结束时间
System.out.println("程序运行时间： " + (endTime - startTime) + "ms");
}
/* 获取数据库连接的函数 */
public static Connection getConnection() {
Connection con = null; // 创建用于连接数据库的Connection对象
try {
Class.forName("com.mysql.jdbc.Driver");// 加载Mysql数据驱动
con = DriverManager.getConnection(
"jdbc:mysql://localhost:3306/local_database", "root",
"034039");// 创建数据连接
} catch (Exception e) {
System.out.println("数据库连接失败" + e.getMessage());
}
return con; // 返回所建立的数据库连接
}
}============END============