Articles & Content


Utilizing Views, RI and Other Stuff for Performance (A08)

Topic: 2008 NA

Subtopic: Cross Platform DB2 for z/OS & LUW

DATE: 2008-05-20 (04:00 PM - 05:00 PM)
SPEAKERS: BL Tysor (Bayard Lee Tysor, Inc.)

The DB2 optimizer has long been recognized for reliably finding the access path for the best response time for most SQL statements. With DB2 V8, this has been taken to new heights. Now with ever decreasing hardware costs, programming cost becomes more and more significant. Codd’s original 12 rules for relational data stressed the goal of separating program logic from specific data implementation. With the advent of OO Programming, and SOA, this becomes even more desirable.Now with DB2 version 8, that separation can be further realized utilizing views. Views can make programming easier and more productive, while allowing for the changing of the underlying data organization. This session will cover how to set up Views, RI and etc. so that DB2 can adjust the access path to the underlying data based on predicate data both at bind time and execution time. Taking advantage of these new capabilities will make application programmers more productive while enhancing query response time.

EXP. LEVEL: Beginner,Intermediate,Advanced


Data Matters: The objective this section is for attendees to be able to list different ways that data can be split up ("sliced and diced") so that the optimizer will be able to minimize the search necessary to retrieve all qualifying rows.

Views for Domains: In this section the objective is for the attendees to be able to describe how to use Views to hide the exact definition of the domain of data from the Application, and why hiding the exact definition can be advantageous.

Layering Views: In this section the objective is for the attendees to be able to layer Views to isolate Applications from the definitions of domains or the underlying tables, there by isolate them from changes to those domains and/or base tables.

Applying Views: In this section the objective is for the attendees to understand how views can be used for Easier Programming, Logical Data Independence, Enhancing Performance, and Reuse.

Strategies for Performance: In this section, the objective is for the attendees to be able to utilize RI, Table Constraints, Unique Indexes, etc. to influence the optimizer in selecting the best access path, especially when utilizing Views.

This file is being provided by IDUG. We would encourage you to join IDUG to get full access to all of our files and resources. Joining IDUG is FREE and signing up is simple. Click here to join! or login!

Download File
Click to Download

NOTE: These are only open to members of IDUG. If you are not a member, please CLICK HERE for more information.