The Challenge: The “Right” Design in Database Management

In the world of database design, “getting it right” is a phrase filled with ambiguity, for who can say what “right” truly is? “Right,” when a legacy system was first created, might not be adequate in today’s rapidly-moving, Internet-based world. To complicate matters, there might not be a way to make “wrong” work in such a way that it’s effectively “right” for today’s situation.

Frankly, it’s a lot easier to spot “bad” or “wrong” database schemas than it is to confirm that a database is “right”. “Right” means that the database is working fine, productivity is high, performance is optimal, the corporate mission is being supported, the organization has the flexibility and underlying foundation, as expressed in the database schema, to move in directions that it needs to move in.

“Wrong” can mean everything from operational constraints that are inhibiting corporate growth and development to partial transactions being applied to the database. When I’m called in to investigate a “wrong” condition, I try to get a holistic view of the situation. While schema design is the foundation of a well-performing database, I would be missing opportunities if I didn’t take database performance, administration and backup/recovery and disaster recovery schemes into consideration.

Each project and situation is going to be a little different from anything that you’ve experienced before, so when starting on a new project, do your homework. Research the corporate goals, strategic business initiatives and underlying business requirements. Understand how database management processes tie the database design to performance tuning and administration and even backup and recovery. Integrate the best schema design that you can after researching current best practices in performance management; ensure that the deployed SQL code is properly tuned and have a plan to keep it maintained. This will go a very long way toward insuring that the database environment you’re working with is functioning at an optimal level.

In the process, try to ensure that the data architects, the DBAs and the data center managers are aligned and working together efficiently.