Editorials

Managing Database Change

DBASchool – 3 Days of In-Person Training
Stephen Wynkoop (SSWUG.ORG Founder/Editor!) will be presenting an exclusive three-day course on what you need to know about SQL Server for your every day support of your systems. The class has very limited seats available and will focus on real-world instruction, discussion about how it applies specifically to your systems and focus on a plan you can take back and apply immediately. From security to performance to best practices and much more – check out the course site and make sure you reserve one of the last remaining seats ASAP.

[Get More Information] or [Register Here]

Managing Database Change
From Ben Taylor
I am constantly reminded of the need to manage change in a database. Changing code in an application is difficult enough. There are many tools that help manage code change in a pretty straight forward fashion. As developers, we are comfortable with change and source control. Most likely, these tools are already integrated into your development tools. It is the rare organization that edits code in a production system, and this is considered to be a very bad practice.

But a database is different simply because it is not only code; it also contains data. In fact, it may be the data itself that is being modified. It may be stored procedures or functions that depend on schema changes. There are many vendors who have great tools helping to manage this process. Many compare a before image of a database to an after image, and generate SQL scripts to transform the former to the latter. Some tools manage the change from a data model, or master repository, generating the necessary change scripts predicting what the necessary change to the database.

Most of these methods don’t work at a granular level. If you have two different initiatives being developed against a single Database and only one initiative is being shipped to production, how do you apply the changes for the single initiative and ignore the changes for the other? It can be done. It’s just harder to get the right package, and the process is a manual one sometimes results in human error. How about some feedback? Drop me a note and let me know how you solve this problem. Send your responses/questions/ideas to btaylor@sswug.org.

Featured Article(s)
Understanding the Power of Oracle SELECT Statement: Part III
In this article we shall see how to place the SELECT statement within the context of the code and how to use SELECT and JOIN and conditional JOIN to extract information from two or more tables.