数据库连接池DBCP浅析

一个典型的关系数据库应用:
a. 建立连接
b. 执行数据库操作
c. 关闭连接

其中建立连接可能需要数ms才能完成。对于多并发请求的场景,频繁的建立连接、断开连接可能会成为程序与关系数据库交互的瓶颈。
为此,很多时候,我们使用数据库连接池来复用连接,避免频繁的建立连接带来的资源浪费。
DBCP是应用非常广泛的关系数据库连接池,和DBCP类似的有C3p0, Proxool

DBCP依赖commons-pool提供的连接池,其包括两个不同的版本:
a. DBCP 1.4只能运行在JDK1.6(JDBC 4)
b. DBCP 1.3只能运行在JDK 1.4-1.5(JDBC 3)


DBCP相关配置:DBCP configuration
说明:
dataSource: 要连接的 datasource (通常我们不会定义在 server.xml)
defaultAutoCommit: 对于事务是否 autoCommit, 默认值为 true
defaultReadOnly: 对于数据库是否只能读取, 默认值为 false
driverClassName:连接数据库所用的 JDBC Driver Class,
maxActive: 可以从对象池中取出的对象最大个数,为0则表示没有限制,默认为8
maxIdle: 最大等待连接中的数量,设 0 为没有限制 (对象池中对象最大个数)
minIdle:对象池中对象最小个数
maxWait: 最大等待秒数, 单位为 ms, 超过时间会丟出错误信息
password: 登陆数据库所用的密码
url: 连接数据库的 URL
username: 登陆数据库所用的帐号
validationQuery: 验证连接是否成功, SQL SELECT 指令至少要返回一行
removeAbandoned: 是否自我中断, 默认是 false
removeAbandonedTimeout: 几秒后会自我中断, removeAbandoned 必须为 true
logAbandoned: 是否记录中断事件, 默认为 false
minEvictableIdleTimeMillis:大于0 ,进行连接空闲时间判断,或为0,对空闲的连接不进行验证;默认30分钟
timeBetweenEvictionRunsMillis:失效检查线程运行时间间隔,如果小于等于0,不会启动检查线程,默认-1
testOnBorrow:取得对象时是否进行验证,检查对象是否有效,默认为false
testOnReturn:返回对象时是否进行验证,检查对象是否有效,默认为false
testWhileIdle:空闲时是否进行验证,检查对象是否有效,默认为false
initialSize:初始化线程数

DBCP原理
用户使用数据库连接池流程:
a. 从连接池中获取一个连接(如果有已建立空闲的连接,直接获取连接,否则建立新连接)
b. 执行数据库操作
c. 将连接归还给数据库连接池

源码分析(common-dbcp 1.4)

BasicDataSource是用户使用的类,DBCP项目组放出example:
import javax.sql.DataSource;
import java.sql.Connection;
import java.sql.Statement;
import java.sql.ResultSet;
import java.sql.SQLException;


import org.apache.commons.dbcp2.BasicDataSource;


public class BasicDataSourceExample {

    public static void main(String[] args) {
        // First we set up the BasicDataSource.
        // Normally this would be handled auto-magically by
        // an external configuration, but in this example we'll
        // do it manually.
        //
        System.out.println("Setting up data source.");
        DataSource dataSource = setupDataSource(args[0]);
        System.out.println("Done.");

        //
        // Now, we can use JDBC DataSource as we normally would.
        //
        Connection conn = null;
        Statement stmt = null;
        ResultSet rset = null;

        try {
            System.out.println("Creating connection.");
            conn = dataSource.getConnection();
            System.out.println("Creating statement.");
            stmt = conn.createStatement();
            System.out.println("Executing statement.");
            rset = stmt.executeQuery(args[1]);
            System.out.println("Results:");
            int numcols = rset.getMetaData().getColumnCount();
            while(rset.next()) {
                for(int i=1;i<=numcols;i++) {
                    System.out.print("\t" + rset.getString(i));
                }
                System.out.println("");
            }
        } catch(SQLException e) {
            e.printStackTrace();
        } finally {
            try { if (rset != null) rset.close(); } catch(Exception e) { }
            try { if (stmt != null) stmt.close(); } catch(Exception e) { }
            try { if (conn != null) conn.close(); } catch(Exception e) { }
        }
    }

    public static DataSource setupDataSource(String connectURI) {
        BasicDataSource ds = new BasicDataSource();
        ds.setDriverClassName("oracle.jdbc.driver.OracleDriver");
        ds.setUsername("scott");
        ds.setPassword("tiger");
        ds.setUrl(connectURI);
        return ds;
    }

    public static void printDataSourceStats(DataSource ds) {
        BasicDataSource bds = (BasicDataSource) ds;
        System.out.println("NumActive: " + bds.getNumActive());
        System.out.println("NumIdle: " + bds.getNumIdle());
    }

    public static void shutdownDataSource(DataSource ds) throws SQLException {
        BasicDataSource bds = (BasicDataSource) ds;
        bds.close();
    }
}

This ad is supporting your extension Smooth GesturesTurn off these ads?
用户在new 一个BasicDataSource时,并未真正建立连接池,建立连接是在用户首次使用getConnection获取连接时发生。
getConnection调用CreateDataSouce获取DataSource,其用synchronized修饰,保证同一时刻最多只有一个线程执行该段代码

如下:

public Connection getConnection() throws SQLException {
        return createDataSource().getConnection();
    }
protected synchronized DataSource createDataSource()
        throws SQLException {
        if (closed) {
            throw new SQLException("Data source is closed");
        }

        // Return the pool if we have already created it
        if (dataSource != null) {
            return (dataSource);
        }

        //  用户第一次使用getConnection请求连接时,建立数据库连接池
        ConnectionFactory driverConnectionFactory = createConnectionFactory();

        //建立连接池,使用commons-pool的GenericObjectPool
        createConnectionPool();

        // Set up statement pool, if desired
        GenericKeyedObjectPoolFactory statementPoolFactory = null;
        if (isPoolPreparedStatements()) {
            statementPoolFactory = new GenericKeyedObjectPoolFactory(null,
                        -1, // unlimited maxActive (per key)
                        GenericKeyedObjectPool.WHEN_EXHAUSTED_FAIL,
                        0, // maxWait
                        1, // maxIdle (per key)
                        maxOpenPreparedStatements);
        }

        // 设置连接池工厂
        createPoolableConnectionFactory(driverConnectionFactory, statementPoolFactory, abandonedConfig);

        // 建立数据库连接池实例
        createDataSourceInstance();
        
       // 根据配置,初始化建立一些数据库连接
        try {
            for (int i = 0 ; i < initialSize ; i++) {
                connectionPool.addObject();
            }
        } catch (Exception e) {
            throw new SQLNestedException("Error preloading the connection pool", e);
        }
        
        return dataSource;
    }
protected void createDataSourceInstance() throws SQLException {
        PoolingDataSource pds = new PoolingDataSource(connectionPool);
        pds.setAccessToUnderlyingConnectionAllowed(isAccessToUnderlyingConnectionAllowed());
        pds.setLogWriter(logWriter);
        dataSource = pds;
    }


PoolingDataSource封装数据库连接池,其实现了getConnection,如下:

public Connection getConnection() throws SQLException {
        try {
            Connection conn = (Connection)(_pool.borrowObject());
            if (conn != null) {
                conn = new PoolGuardConnectionWrapper(conn);
            } 
            return conn;
        } catch(SQLException e) {
            throw e;
        } catch(NoSuchElementException e) {
            throw new SQLNestedException("Cannot get a connection, pool error " + e.getMessage(), e);
        } catch(RuntimeException e) {
            throw e;
        } catch(Exception e) {
            throw new SQLNestedException("Cannot get a connection, general error", e);
        }
    }

PoolableObjectFactory是生成连接池对象的工厂,其实现了makeObject,如下:

public Object makeObject() throws Exception {
        Connection conn = _connFactory.createConnection();
        if (conn == null) {
            throw new IllegalStateException("Connection factory returned null from createConnection");
        }
        initializeConnection(conn);
        if(null != _stmtPoolFactory) {
            KeyedObjectPool stmtpool = _stmtPoolFactory.createPool();
            conn = new PoolingConnection(conn,stmtpool);
            stmtpool.setFactory((PoolingConnection)conn);
        }
        return new PoolableConnection(conn,_pool,_config);
    }

PoolableConnection是连接代理,其实现了close,close并不真正关闭连接,而是调用_pool.returnObject(this)将连接还给对象池:

public synchronized void close() throws SQLException {
        if (_closed) {
            // already closed
            return;
        }
        ......
        if (!isUnderlyingConectionClosed) {
            // Normal close: underlying connection is still open, so we
            // simply need to return this proxy to the pool
            try {
                _pool.returnObject(this); // XXX should be guarded to happen at most once
            } catch(IllegalStateException e) {
         ......
    }
上述borrowObject和returnObject都是GenericObjectPool提供的接口,分别用来从对象池中获取对象和归还对象。

getConnection序列图如下:
数据库连接池DBCP浅析_第1张图片


其它开源数据库连接池:C3P0、BoneCP、Proxool、阿里巴巴的Druid等
一些性能比较的文档:
c3p0、dbcp、proxool、BoneCP比较 
DBCP,C3P0,Tomcat_JDBC 性能及稳定性测试
各种数据库连接池对比

reference:
commons-dbcp

你可能感兴趣的:(数据库连接池DBCP浅析)