IT Management

Database Administration Best Practices

5 Pages

How can data professionals who are stuck in “react” mode get out of their rut? Develop a list of best practices, tie them to enterprise strategies and departmental goals, and grade them on a quality scale (like that of the Capability Maturity Model). Such a plan is needed to provide continuous improvement.


I am a DBA in an IBM server environment (the hardware and database management system aren’t relevant here). I’m no longer on pager rotation, so I’m able to relax on weekends. As I dive into my pizza and root beer, with some classical music playing in the background, I ponder my reason for existence, stated thusly: What did I do this week, and why did I do those things?

The following realities strike me:

  • Every time someone asks me what I do for a living, I have a difficult time explaining it. “I’m a database administrator” usually results in blank stares.
  • During my normal workday, I spend my time reacting to things; most of these things may be symptoms of underlying problems, but I don’t have time to figure those out. I’m paid to get things done, not to fix things.
  • I have difficulty organizing and prioritizing my work in the current environment. I must react to whatever is “hot,” not the day-to-day tasks.
  • I don’t have time to plan!

You probably have heard most of these by now. Many of these statements have stock answers such as, “If you fail to plan, you are planning to fail!” or “Transform your environment from a reactive one to a proactive one.” Specific to the DBA, various industry experts and authors have attempted to explain what DBAs should do and why they should perform various tasks. Unfortunately, there is a general lack of understanding in the industry of what a DBA does and why it needs to be done. DBAs need ways to:

  • Determine the best practices to use when implementing enterprise strategies and departmental goals
  • Measure the quality of current documentation, processes, and practices
  • Develop processes for continuous quality improvement
  • Craft project plans and prioritize tasks.

This is a lot to ask, but I believe it is critical. Without this, we are stuck in the rut of reacting to the “Hot Symptom” of the moment and we’ll never get out. This article presents a plan for getting yourself and your department out of this rut.


Have we learned anything in almost three decades of database administration? Have enterprises developed proactive strategies for enterprise data administration, or are many still lost in the wilderness of reacting to the symptom(s) of the day?

I’ve talked at great length with other DBAs. Most of us wish we could be proactive, but it just isn’t happening. Management forces us into react mode, and the current economic downturn and rising healthcare costs (among other things) are forcing shops to do more with less.

Such a reactive strategy also leads to problem solving using a “band-aid” approach: If a quick fix is available, it goes in! Too many IT organizations throw CPU horsepower, DASD storage, or memory at symptoms, hoping they will go away. Similarly, too many DBAs throw buffer pools, indexes, and DASD at poorly performing applications, thereby either neglecting or obscuring the underlying problems.

5 Pages