As companies collect ever more data, there’s a sense that the role of the database administrator is changing, but there’s little consensus on exactly how.
"One of the biggest concerns I keep hearing is how the DBA role has either lost importance or gained importance. There’s been a change and there’s disagreement about what the change is," Bert Scalzo, database domain expert for Quest Software, told me for an article at Enterprise Apps Today.
Some view the role as becoming more important as companies keep more data live longer, yet don’t hire more DBAs to deal with it. Other companies – especially those using cloud services – consider that part of the service, so they don’t think they need DBAs for tuning, optimization, backup and recovery, he said.
With new technologies, specific skills with Linux, Windows, HP-UX or AIX become less important, he said. The real trick to making the transition is keeping an open mind:
The biggest thing [in making the transition] is not to presuppose or apply your Golden Rules that you’ve learned over your last 10 or 20 years. Good database design in a relational environment might not be good design in a column-oriented data store. How you define data structures relationally might be different than how you define them in a noSQL database. It’s kind of an unlearning/learning process that you have to embrace. You still apply common sense; you just use you experience to gauge that common sense in a new paradigm.
Gwen Shapira, senior database administrator at data infrastructure management company Pythian, who wrote a long piece on how DBAs are prime candidates for a new role called Hadoop cluster administrator, told me she sees the move to Big Data as a huge opportunity for DBAs:
"In the age of Big Data, data professionals are more visible to the organization than ever, and they are seen as the critical resource they always were."
"Talk to business unit managers to find their pain points. Which information do they need? …Managers are happier to talk to technical people now than they've ever been. Show them how your skills can help them get more value out of data, because this is what Big Data means to them."