Improving SQL Performance

Improving SQL Performance

by: Marisa Pellegrino

How do you know how much hardware is really needed by your applications? And what do you do when your applications are overloading your system? The answer lies with improving your SQL performance. You have to tune your hardware SQL server and monitor performance, all of which will be explained as clearly as possible on this page.

The first thing to do when you want to improve your SQL performance is you need to learn how to optimize your system by finding out how much hardware you really need to run your applications. The best way to tune your hardware and monitor performance is through the art of performance monitoring which takes experience, knowledge, and sometimes even luck.

Performance monitoring guidelines:

Make sure you’re running your typical processes and work loads during the monitoring.

Don’t only do a realtime monitoring of your servers; capture long running logs.

Always have the disk counters turned ON.

Set up the chart windows with an interval of 18 seconds for routine, daily desktop monitoring.

Know the tools you are working with.

Don’t be afraid to experiment.

Know the terminology (ขobjectsข are lists of individual stats available; ขcounterข is a single stat; ขinstanceข is further breakdown of a counter stat into duplicate components).

A bottleneck happens when the hardware resources can’t keep up with the demands of the software. This is usually fixed in one of two ways: first, you identify the limiting hardware and increase its potential (i.e. a faster hard drive or increase the speed of the computer); second, make the software processes use the hardware more efficiently.

Five areas to watch when improving SQL performance and identifying bottlenecks:

Memory usage

CPU processor utilization

Disk input/output performance

User connections

Blocking locks

About The Author

Marisa Pellegrino is a freelance writer from Montreal and is the head researcher and content manager for www.sqlrecovery.com.

[email protected]

This article was posted on December 24, 2004

by Marisa Pellegrino