Duration
4 days
Course Description
This classroom course explains how to monitor and tune DB2 9 for z/OS from a system point of view.
Audience
This course is for DB2 system tuners, Enterprise capacity planners and System programmers.
Pre-requisites
You should have:
- DB2 9 for z/OS Database Administration Workshop Part 1 (CV830MY), DB2 9 for z/OS Database Administration Workshop Part 1 - Instructor Led Online (3V830), or equivalent experience
- DB2 9 for z/OS Database Administration Workshop Part 2 (CV840), DB2 9 for z/OS Database Administration Workshop Part 2 - Instructor Led Online (3V840), or equivalent experience
- DB2 9 for z/OS System Administration (CV850MY), DB2 9 for z/OS System Administration - Instructor Led Online (3V850), or equivalent experience
Course Objective
- Describe DB2 processes as they relate to DB2 performance
- Interpret DB2 Trace Output
- Tune a DB2 subsystem
- Provide good input to system programmers to allow them to effectively tune the z/OS system for best DB2 performance
- These objectives are met by discussing how to:
- Monitor and control the DB2 workload
- Monitor and tune IMS/CICS/DDF performance
- Monitor and tune DB2 buffer pools, EDM pool, logging and IRLM execution
- Tune DB2 attachments, including the DB2 side of distributed requests
- Analyze bottlenecks
- Communicate with systems people about DB2-related issues
- Communicate with application developers on DB2 performance topics
- Interpret DB2 statistics and accounting traces
- List guidelines which enhance DB2 system performance
- Understand how DB2 works well enough to be able to react effectively to many situations
Key topics
DB2 system performance analysis and tuning introduction
- Identify why some z/OS concepts and controls have important influences on DB2 performance
- Learn how DB2 works to enable effective monitoring and tuning of DB2
The DB2 trace
- Set up and handle DB2 traces for both continuous monitoring and problem analysis
- Determine, depending on the situation, whether to start with Statistics or with Accounting reports
- Decide how to monitor and tune in your environment
- List basic principles of reading and understanding monitor reports
Introduction to the operating system and hardware
- Discuss operating system and hardware topics with other systems people
- Identify the features and limits of specified hardware
- Contrast reports originating from DB2 with those originating from the operating system
Service time, queuing time, case A transaction analysis
- Analyze elapsed time components reported by accounting trace
- Service versus queuing
- Dominating components
- Tuning potential
- Analyze capacity bottlenecks with basic queuing theory
- Estimate effect of software setup and hardware alternatives on elapsed time
Introduction and analysis of case E
- Do an in-depth analysis of an accounting trace
The DB2 attachments
- Obtain accounting information from the different environments (TSO, CICS, IMS, and so forth)
- Recognize accounting anomalies that may occur
- Take measures to increase/decrease the maximum amount of work that can flow into DB2
- Take measures to prevent query users from overloading your system
- Apply performance options in the CICS, IMS and Distributed interfaces to DB2
In-depth application of the DB2 system
- Monitor current usage (behavior) of the most important parts of DB2
- Differentiate between good and bad behavior
- Perform DB2 system level tuning
- Assist application developers in exploiting selected DB2 system level features
- Identify things application developers should not do
The disk subsystems and DB2 I/O
- Identify how disk I/O works for DB2 operations
Capacity analysis and massive batch
- Predict the effect of tuning actions and hardware changes
- Identify the issues of running massive batch jobs in parallel with performance-critical transactions
Programme scope, delivery arrangements and suitability are confirmed during enquiry.