Performance Tuning of a Daffodil DB / One$DB -JDBC Application


This article illustrates the best practices to improve the performance of Daffodil DB / One$DB JDBC Driver. This article focuses on how to improve the performance of a Daffodil DB / One$DB JDBC application using Statement, PreparedStatemnt, CallableStatement and ResultSet interfaces. Choosing the right statement interfaces and right methods according to your SQL query plays a vital role in improving the performance of a JDBC Driver.

JDBC Overview

JDBC API provides standard set of interfaces to work with databases like Daffodil DB / One$DB, Oracle, Derby etc.

Connection interface encapsulates database connection functionality, Statement interface encapsulates SQL statement representation and execution functionality whereas ResultSet interface encapsulates retrieving data which comes from the execution of a SQL query using Statement.

Following are the basic steps to write a JDBC program.

1. Import "java.sql" and "javax.sql" packages. (Import "javax.sql", if advanced JDBC feature like XA is to be used)

2. Load Daffodil DB JDBC driver (embedded or network JDBC Driver)

3. Establish connection to database using Connection interface

4. Create a Statement

5. Execute the Statement

6. Retrieve results by using ResultSet interface

7. Close Statement and Connection

Choosing right Statement interface:

There are three types of Statement interfaces in JDBC to represent/execute a SQL query-Statement, PreparedStatement and CallableStatement. Statement is used for executing static SQL statement with no input and output parameters; PreparedStatement is used to execute dynamic SQL statement with input parameters whereas CallableStatement is used to execute dynamic SQL with both input and output parameters. One important thing to note about PreparedStatement and CallableStatement is that they can also be used for static SQL statements. However, CallableStatement is mainly meant for stored procedures.

PreparedStatement gives better performance when compared to Statement because it is pre-parsed and pre-compiled. This means that compilation and parsing of such statement is done only once by the database. Afterwards the database reuses the already parsed and compiled statement. This significantly improves the performance because whenever a statement has to be executed repeatedly, it doesn't need to be parsed and compiled time and again. So the overload incurred by parsing and compiling the same statement can be reduced.

When there is a requirement for single request to process multiple complex statements, CallableStatement gives better performance as compared to PreparedStatement and Statement.

To read the full article please visit http://www.daffodildb.com/daffodildb-performance-tuning.html

This article has been contributed by (Mr.) Parveen Aggarwal, Technical Consultant to DSL India (http://www.daffodildb.com). With more than 6 years of industry experience in Java and allied technologies, he has an in-depth understanding of J2EE, J2ME and database management systems. Parveen is currently working on the concept of data-archiving in embedded databases. He can be contacted at [email protected]

home | site map
© 2005