使用JDBC连接到数据库的时间

Question

When I test the run time like this,the time on my machine is 735

 public class TestDBCP {

    static
    {
        try {
            Class.forName("com.mysql.jdbc.Driver");
        } catch (ClassNotFoundException e) {
            // TODO Auto-generated catch block
            e.printStackTrace();
        }
    }
    public static void main(String[] args) {
    long l = System.currentTimeMillis();
    Connection conn = null;
    try {
        for(int i =0; i < 100; i++)
        {
            conn = DriverManager.getConnection("jdbc:mysql://localhost:3306/jdbcdemo","root","Sph815@cs");
            conn.close();
        }
    } catch (SQLException e) {
        // TODO Auto-generated catch block
        e.printStackTrace();
    }
    System.out.println(System.currentTimeMillis() - l);
    }

     }

But when I test it using the following its run time becomes longer, i don't understand why more job fewer time?

public class TestDBCP {

static
{
    try {
        Class.forName("com.mysql.jdbc.Driver");
    } catch (ClassNotFoundException e) {
        // TODO Auto-generated catch block
        e.printStackTrace();
    }
}
public static void main(String[] args) {
long l = System.currentTimeMillis();
    try {
        for(int i =0; i < 100; i++)
        {
 DriverManager.getConnection("jdbc:mysql://localhost:3306/jdbcdemo","root","Sph815@cs");
        }
    } catch (SQLException e) {
        // TODO Auto-generated catch block
        e.printStackTrace();
    }
    System.out.println(System.currentTimeMillis() - l);
    }

}

Answer 1

You're not closing the connection immediately in the second one. Opening and closing one connection 100 times won't take nearly as long as opening 100 connections at once.

Answer 2

You are assuming that execute an extra statement (the one that close the connection) will implies that the program will take more time because the close() statement execution adds extra CPU usage.

If you doesn't execute it, you save this CPU time, however the database has to deal with more open connections simultaneously , which can degrade his performance. Therefore, the getConnection() statement will takes longer to execute as the number of open connections increases.

Probably this is the reason your second version execute slower than the first one.