When we talk about the benefits of using a non-relational database management system, often referred to as a NoSQL database, we sometimes lose track of what a traditional database is still good for (for some background on what a relational database is, see our guide to data terminology).

In a blog post at DBMS2, database veteran Curt Monash explains when it’s still best to use a relational database.

Monash gives the following use-cases as examples of when a relational database is best:

  • You’re building a low-volume, medium-complexity suite of applications that will evolve over time.
  • Your (duplicated) data volumes would be ridiculous if you didn’t do a reasonable amount of normalization.
  • You simply don’t see a cost/benefit advantage to moving away from proven legacy technology.

Monash also gives recommendations on when it’s less clear whether one should use a relational or non-relational database and notes that various next-generation relational databases (which have been called NewSQL) may change the equation in the future.

klint finley