practical_title

Identifying the cause of the problem accurately is
more important than solving the problem!

Database performance issues arise in various forms, caused by various factors, in various environments; and MaxGauge provides various functions to monitor and analyze such issues. The Practical Troubleshooting deals with issues that occur more frequently in general, among various kinds of performance issues. Especially, it provides a process of analyzing performance issues caused by situations such as the batch program or batch query which have been performing normally, not performing within the batch window, and competitions such as lock.
We hope the following case studies will help you to become familiar with MaxGauge’s performance analysis methodology.

 

case_2_thumb

Case1. Analyzing Performance Degradation Caused by Changes in Execution Plan

Like the recent, it is a clear fact that the CBO (Cost Based Optimizer) is a more suitable optimizer in an environment where a large amount of data is accessed under various conditions by many users.  However, one of the problems of CBO is that the execution plan could change unexpectedly.

case_1_thumb

Case2. Analyzing Performance Degradation caused by DOP (Degree of Parallelism) Downgrade

The database nowadays not stores the DW (Dataware House), ODS (Operational Data Store), but also the OLTP system, or a large data as well.  The major functions which the Oracle provides in order to process a large data effectively are partitioning and parallel processing.

case_3_thumb

Case3. How to Eliminate Potential Risk Factors through the  Analysis of Pro-active Lock Issues

The Oracle ensures data consistency through the lock mechanism.  This is very essential in terms of transaction protection.  However, when multiple users are trying to change the same data, there is a high chance of generating a lock wait phenomenon.  Therefore, in order to minimize the lock wait time, you must reduce the transaction process time as much as possible.