1. Definition
C3P0 is an open-source JDBC connection pool, which implements the data source and JNDI binding, supports the JDBC3 specification and implements the Connection and Statement pool DataSources objects of the JDBC2 standard extension description.
The connections used to connect to the database are integrated to form an on-the-go database connection pool (Connection pool).
2. Why use C3P0 database connection pool
2.1. Huge consumption of time and memory resources
Use JDBC legacy mode
public static void main(String[] args) {
BaseDao2<Dog> dao = new BaseDao2<>();
System.out.println(Timestamp.valueOf(LocalDateTime.now()));
for(int i=0;i<100;i++){
Connection connection=dao.getConnection();
try {
connection.close();
} catch (SQLException e) {
e.printStackTrace();
}
}
System.out.println(Timestamp.valueOf(LocalDateTime.now()));
}
Connect using C3P0
public static void main(String[] args) {
BaseDao2<Dog> dao = new BaseDao2<>();
System.out.println(Timestamp.valueOf(LocalDateTime.now()));
for(int i=0;i<100;i++){
Connection connection = getConnection();
try {
connection.close();
} catch (SQLException e) {
e.printStackTrace();
}
}
System.out.println(Timestamp.valueOf(LocalDateTime.now()));
} 
100 connections have been made, using JDBC traditional mode connection, it takes nearly 2s
When using the C3P0 database connection pool, it takes less than 1 second.
When the amount of data is large, the time-consuming gap will be more obvious.
This happens because:
Ordinary JDBC database connections are obtained using DriverManager. Every time a connection is established to the database, the Connection must be loaded into memory, and then its correctness is verified according to the username and password in the JDBC code (or configuration file ) .
This process generally takes 0.05~1s. Once a database connection is required, one must be requested from the database, and the connection is disconnected after execution.
If dozens or even hundreds of people request to connect to the same database at the same time, it will inevitably take up a lot of system resources, and seriously cause the server to crash.
2.2. There is a risk of memory leaks
Every time the database connection is used, it needs to be disconnected, but if the program fails to close the connection in time due to an abnormality, this may cause a memory leak , which can only be solved by restarting the database in the end;
In addition, the development using the traditional JDBC mode cannot control the number of connections that need to be created . The system generally allocates a large number of resources to connections to prevent insufficient resources. If the number of connections exceeds a certain number, it may cause memory leaks.
2.3. The application directly obtains the link
shortcoming:
1. The user needs to obtain a link to the database for each request, and the creation of a connection to the database usually consumes relatively large resources and takes a long time to create
2. Assuming that the website has 100,000 visits a day, the database server needs to create 100,000 connections, which greatly wastes database resources and easily causes memory overflow and downtime of the database server
2.4, use the database connection pool to optimize
Three, C3P0 practical operation
3.1. Import jar package
<dependency>
<groupId>com.mchange</groupId>
<artifactId>c3p0</artifactId>
<version>0.9.5.2</version>
</dependency>3.2, configuration xml file
The following is the configuration xml file, which can be modified according to the actual situation.
3.3, c3p0-config.xml template
<c3p0-config>
<named-config name="mysqlapp">
<property name="driverClass">com.mysql.cj.jdbc.Driver</property>
<property name="jdbcUrl">jdbc:mysql://192.168.153.134:3306/new</property>
<property name="user">root</property>
<property name="password">123123</property>
<!-- 进行数据库连接池管理的基本信息 -->
<!-- 当数据库连接池中的连接数不够时，c3p0一次性向数据库服务器申请的连接数 -->
<property name="acquireIncrement">5</property>
<!-- c3p0数据库连接池中初始化时的连接数 -->
<property name="initialPoolSize">10</property>
<!-- c3p0数据库连接池维护的最少连接数 -->
<property name="minPoolSize">10</property>
<!-- c3p0数据库连接池维护的最多的连接数 -->
<property name="maxPoolSize">200</property>
<!-- c3p0数据库连接池最多维护的Statement的个数 -->
<property name="maxStatements">50</property>
<!-- 每个连接中可以最多使用的Statement的个数 -->
<property name="maxStatementsPerConnection">2</property>
</named-config>
</c3p0-config>3.4, c3p0-config.xml parameter list
<c3p0-config>
<default-config>
<!--当连接池中的连接耗尽的时候c3p0一次同时获取的连接数。Default: 3 -->
<property name="acquireIncrement">3</property>
<!--定义在从数据库获取新连接失败后重复尝试的次数。Default: 30 -->
<property name="acquireRetryAttempts">30</property>
<!--两次连接中间隔时间，单位毫秒。Default: 1000 -->
<property name="acquireRetryDelay">1000</property>
<!--连接关闭时默认将所有未提交的操作回滚。Default: false -->
<property name="autoCommitOnClose">false</property>
<!--c3p0将建一张名为Test的空表，并使用其自带的查询语句进行测试。如果定义了这个参数那么
属性preferredTestQuery将被忽略。你不能在这张Test表上进行任何操作，它将只供c3p0测试
使用。Default: null-->
<property name="automaticTestTable">Test</property>
<!--获取连接失败将会引起所有等待连接池来获取连接的线程抛出异常。但是数据源仍有效
保留，并在下次调用getConnection()的时候继续尝试获取连接。如果设为true，那么在尝试
获取连接失败后该数据源将申明已断开并永久关闭。Default: false-->
<property name="breakAfterAcquireFailure">false</property>
<!--当连接池用完时客户端调用getConnection()后等待获取新连接的时间，超时后将抛出
SQLException,如设为0则无限期等待。单位毫秒。Default: 0 -->
<property name="checkoutTimeout">100</property>
<!--通过实现ConnectionTester或QueryConnectionTester的类来测试连接。类名需制定全路径。
Default: com.mchange.v2.c3p0.impl.DefaultConnectionTester-->
<property name="connectionTesterClassName"></property>
<!--指定c3p0 libraries的路径，如果（通常都是这样）在本地即可获得那么无需设置，默认null即可
Default: null-->
<property name="factoryClassLocation">null</property>
<!--强烈不建议使用该方法，将这个设置为true可能会导致一些微妙而奇怪的bug-->
<property name="forceIgnoreUnresolvedTransactions">false</property>
<!--每60秒检查所有连接池中的空闲连接。Default: 0 -->
<property name="idleConnectionTestPeriod">60</property>
<!--初始化时获取三个连接，取值应在minPoolSize与maxPoolSize之间。Default: 3 -->
<property name="initialPoolSize">3</property>
<!--最大空闲时间,60秒内未使用则连接被丢弃。若为0则永不丢弃。Default: 0 -->
<property name="maxIdleTime">60</property>
<!--连接池中保留的最大连接数。Default: 15 -->
<property name="maxPoolSize">15</property>
<!--JDBC的标准参数，用以控制数据源内加载的PreparedStatements数量。但由于预缓存的statements
属于单个connection而不是整个连接池。所以设置这个参数需要考虑到多方面的因素。
如果maxStatements与maxStatementsPerConnection均为0，则缓存被关闭。Default: 0-->
<property name="maxStatements">100</property>
<!--maxStatementsPerConnection定义了连接池内单个连接所拥有的最大缓存statements数。Default: 0 -->
<property name="maxStatementsPerConnection"></property>
<!--c3p0是异步操作的，缓慢的JDBC操作通过帮助进程完成。扩展这些操作可以有效的提升性能
通过多线程实现多个操作同时被执行。Default: 3-->
<property name="numHelperThreads">3</property>
<!--当用户调用getConnection()时使root用户成为去获取连接的用户。主要用于连接池连接非c3p0
的数据源时。Default: null-->
<property name="overrideDefaultUser">root</property>
<!--与overrideDefaultUser参数对应使用的一个参数。Default: null-->
<property name="overrideDefaultPassword">password</property>
<!--密码。Default: null-->
<property name="password"></property>
<!--定义所有连接测试都执行的测试语句。在使用连接测试的情况下这个一显著提高测试速度。注意：
测试的表必须在初始数据源的时候就存在。Default: null-->
<property name="preferredTestQuery">select id from test where id=1</property>
<!--用户修改系统配置参数执行前最多等待300秒。Default: 300 -->
<property name="propertyCycle">300</property>
<!--因性能消耗大请只在需要的时候使用它。如果设为true那么在每个connection提交的
时候都将校验其有效性。建议使用idleConnectionTestPeriod或automaticTestTable
等方法来提升连接测试的性能。Default: false -->
<property name="testConnectionOnCheckout">false</property>
<!--如果设为true那么在取得连接的同时将校验连接的有效性。Default: false -->
<property name="testConnectionOnCheckin">true</property>
<!--用户名。Default: null-->
<property name="user">root</property>
<!--早期的c3p0版本对JDBC接口采用动态反射代理。在早期版本用途广泛的情况下这个参数
允许用户恢复到动态反射代理以解决不稳定的故障。最新的非反射代理更快并且已经开始
广泛的被使用，所以这个参数未必有用。现在原先的动态反射与新的非反射代理同时受到
支持，但今后可能的版本可能不支持动态反射代理。Default: false-->
<property name="usesTraditionalReflectiveProxies">false</property>
</default-config>
</c3p0-config>3.5. Create C3P0Util class
// C3p0连接池
public class C3p0Util {
private static DataSource dataSource = null;
static{
dataSource = new ComboPooledDataSource("mysqlapp");
}
//从连接池中获取连接
public static Connection getConnection(){
try {
return dataSource.getConnection();
} catch (SQLException e) {
e.printStackTrace(); }
return null;
}
//释放连接回连接池
public static void release(Connection conn, Statement stmt, ResultSet rs) {
if (rs != null) {
try {
rs.close();
} catch (Exception e) {
e.printStackTrace();
}
rs = null;
}
if (stmt != null) {
try {
stmt.close();
} catch (Exception e) {
e.printStackTrace();
}
stmt = null;
}
if (conn != null) {
try {
conn.close();
} catch (Exception e) {
e.printStackTrace();
}
conn = null;
}
}
}