This monograph is a short guide to performance optimisation techniques for the OpenEdge RDBMS. Thought it was written shortly after version 10.1A was released, much of the information in it applies to later OpenEdge versions as well. It is intended to be a useful collection of tips and techniques, not an exhaustive treatise on the subject. Many of the recommendations are quite easy to implement and will provide you significant "bang for the euro" without doing too much work or too much thinking.

Some of the recommendations are much easier to implement if you are creating a new database than if you have an existing database. For example, if you have a large existing database, redesigning the storage layout may be too time consuming to be convenient.

To illustrate what a little performance tuning can easily accomplish, we used a multi-user benchmark with a highly update-intensive workload. When we used an "out of the box" configuration of OpenEdge 9.1D with no tuning, throughput was about 30 transactions per second. By implementing just eight of the suggestions presented here, we obtained over an order of magnitude improvement -- almost 600 transactions per second!

Because every database, computer, operating system configuration, application, and workload is different, there is no universal database configuration that will work best, or even equally well, for everyone. While several of the suggestions presented in this monograph can be useful for most systems, they may not be appropriate for your system. Therefore, when you implement them, you should always make measurements before and after making changes so you can be be sure the effects you get are the ones you wanted. An old carpenter's adage "measure twice, cut once" is apropos -- but the second measurement should come afterward.