DB2 User's point of view: Purequery and DB2 9 sparkle

DB2 9 for z/OS offers a number of new performance features. One of the most exciting feature combinations I have found in DB2 9 and purequery is the ability to convert dynamic JDBC SQL applications into static transactions. This combination uses the new IBM Data Server Driver for JDBC to eliminate the overhead of each JDBC SQL transaction in the context of dynamic JDBC security authorization, object validation, and access path creation. For a DB2 system that executes millions of dynamic JDBC transactions daily, converting a dynamic transaction into a static transaction can save a lot of CPU time and operational overhead.

Purequery further developed the object-oriented nature of DB2. Purequery supports the use of traditional JDBC application SQL programming, JDBC 4.0 SQL method calls, and named query styles in JPA, Hibernate, and Ibatis. By using Purequery, developers can write applications in any programming style and easily convert dynamic applications into static, robust, robust DB2 applications.

With the new eclipse-based programming tools, IBM Data Studio v1.1 uses the improvements in purequery to download the tool free of charge from the IBM website. You can access these features in a Java project perspective with a simple menu item. These features are available by packaging and deploying the Purequery Runtime (Pdqmgmt.jar) and JDBC drivers on any AVA Application server (WebSphere, tomcat, etc.).

DB2 9 also has another impressive feature: the "trusted context" of distributed system transactions. This feature can improve performance while ensuring security. This feature allows you to create a trusted network that combines with database objects and associates an application plan or package with a security database role. Defines a trusted network platform that uses the role as object owner keyword when creating database objects and binds to trusted database roles so that the DB2 system can easily perform security checks and reuse dynamic database transaction lines threads access plans. By caching and simplifying these remote distributed WebSphere Server transactions, DB2 retains the security token as long as the server is still committing transactions to the DB2 for z/OS system. This feature improves security by using tokens, eliminating the overhead of level two security checks for each transaction issued by the server. As I mentioned in my discussion of purequery, this feature can significantly improve performance for systems that handle a large number of distributed dynamic transaction loads.

Finally, DB2 9 supports column-level data encryption. We all need to be very careful about this feature. I recently met an Englishman who had to change his credit card four times because of an information loophole in an organization (including the British government) with whom he had business dealings. These vulnerabilities do not cause problems if these organizations use encryption techniques. The cost of encryption is not large, and encryption can greatly improve security, so that your CEO can be more assured.