Wait Time Analysis

Written by Kimberly Clark
Bookmark and Share

The performance of a database is almost always assessed in terms of response time and throughput. Response time is a description of how long it takes for a single task to execute, whereas throughput is a measurement of the volume of work that is performed in a specified time frame. Thus, to accurately gauge how the database is operating the database administrator (DBA) needs to conduct a detailed wait time analysis on their system.

The process of wait time analysis not only involves determining how long the user or application has to wait for the system to execute their requests. It also includes figuring out why the system stalled in the first place. There are several steps that should be performed to get a realistic picture of the wait events.

Benchmark Database Performance

To start with, the current performance of the database must be benchmarked. The statistical information collected is useful in a couple of ways. First of all, it will assist the DBA in recognizing if the system ever behaves abnormally in the future.

In addition, the information will prove particularly helpful when trying to pinpoint areas that need to be optimized. The statistical information should be comprehensively analyzed to identify those operations that take an exceptionally long time to complete. Furthermore, the reports can really help the DBA zero in on those functions that are hogging the system resources.

Bookmark and Share