1. steps
   1. import the required packages
   2. register the jdbc drivers
      1. Class.forName("com.mysql.jdbc.Driver");
   3. open a connection to a database
      1. DriverManager.getConnection();
   4. create a statement object
   5. execute a query and return a ResultSet object
   6. process the ResultSet object
   7. close the ResultSet and Statement Objcet
   8. close the connnection
2. architecture
   1. jdbc
      1. pic
      2. core
   2. performance
      1. select the right JDBC driver
      2. simplify sql queries
      3. minimizing the use of database metadata methods
      4. manage database connection objects
         1. setting optimal connection properties
         2. setting an optimal row prefetch value
         3. using a connection pool
            1. example
            2. connect once and using multiple statement objects
         4. controlling a transaction
            1. many statement object executed as a single unit
         5. choosing the optimal transaction isolation level
            1. transaction level
            2. TRANSACTION_NONE-0
            3. TRANSACTION_READ_UNCOMMITTED-1
            4. TRANSACTION_READ_COMMITTED-2
            5. TRANSACTION_REPEATABLE_READ-4
            6. TRANSACTION_SERIALIZABLE-8
      5. avoding using generic search patterns
         1. example
      6. retrieving only required data
         1. return the only columns you need
         2. set driver specific methods
            1. Statement.setMaxRows
            2. Statement.setMaxFieldSize
            3. ResultSet.setFetchSize
            4. Statement.setFetchSize
      7. Statement or PrepareStatement
         1. Statement
         2. PrepareStatement
            1. http://www.javaeye.com/topic/49129?page=3
      8. choose the right cursor
         1. ResultSet.TYPE_FORWARD_ONLY
         2. ResultSet.TYPE_SCROLL_INSENSITIVE
         3. ResultSet.TYPE_SCROLL_SENTITIVE
      9. caching PreparedStatement Objects
         1. PreparedStatement object pooling to cache rarely changed sql or queries
   3. tiers
      1. 2 tiers
         1. pic
      2. 3 tiers
         1. pic
3. API
   1. java.sql
      1. elements
         1. DriverManager
            1. driver
            2. jdbc-odbc
            3. pic
            4. not recommended
            5. experimental
            6. not as fast as jdbc
            7. do not support multiple concurrent open statements
            8. not support Blob and Clob
            9. install
            10. how to load
            11. Class.forName("com.mysql.jdbc.Driver");
            12. DriverManager.registerDriver(new com.mysql.jdbc.Driver());
            13. instance a Driver class
            14. use System.properties()
            15. java -Djdbc.drivers=com.mysql.jdbc.Driver:sun.jdbc.odbc.JdbcOdbcDriver Test
            16. use a thread class
            17. type
            18. JDBC-ODBC
            19. some ODBC native code must be loaded on each client machine
            20. pic
            21. native-API
            22. requires some binary code be loaded on each client machine
            23. pic
            24. network-protocol
            25. translate JDBC API calls into a DBMS-independent net protocol
            26. pic
            27. native-protocol
            28. convert JDBC calls into the network protocol used by DBMS directly
            29. pic
            30. select criteria
            31. features and ease of use
            32. speed/performance
            33. reliability
            34. security
            35. protability
            36. support
            37. price
            38. open source
            39. compatibility
         2. Connection
            1. URL
            2. jdbc:mysql://localhost/db1
            3. pic
            4. Class.forName()
            5. unicode
            6. datasource
            7. JNDI
            8. PIC
            9. benefits
            10. properties
            11. pic
            12. create DataSource Object
         3. Statement
            1. execute()
            2. executeBatch()
            3. executeQuery()
            4. executeUpdate()
         4. PreparedStatement
            1. example
         5. CallableStatement
         6. ResultSet
            1. get index start from 1
            2. get table row count
         7. SQLExcepiton
   2. javax.sql
      1. DataSource to establish a connection
      2. Connection pooling
         1. benefits
            1. limited connections
            2. performance
         2. interface
            1. ConnectionPoolDataSource
            2. PooledConnection
         3. products
            1. Apache Avalon/Excalibur
            2. Commons Pool
            3. commons DBCP
      3. Distributed transactions
      4. Row sets
   3. metadata
      1. DatabaseMetaData
      2. ResultSetMetaData
      3. ParameterMetaData
      4. RowSetMetaData
   4. transaction
      1. default autocommit is true
         1. sql statements are executed and committed as individual transaction
         2. for every single statement, a transaction begins-->execute the statement successfully-->commit
      2. conn.setAutoCommit(false);
      3. rollback
         1. any point before committing a transaction
         2. conn.rollback();
      4. how to start
         1. explicit
         2. then set autocommit back to true
   5. debug
      1. DriverManager.setLogWriter(pw);
4. special
   1. data type
      1. Blob
      2. Clob
      3. Date/Time/Timestamp