The Evolving DBA The world is changing. Plenty of people have written about the changes in the job of the DBA (DBAKevlar, Brent Ozar), including myself. While lots of vendors might postulate their database removes the need for a DBA, and some might believe that, I think that's disingenuous. Microsoft doesn't hire DBAs, but they hide that job under other titles. We still need DBAs, though the job changes a little in the future. Recently I saw a post from new PASS Board director, Melody Zacharias, where she notes that DBAs are more important than ever. She gives us five reasons, and I tend to agree with most of them. There is a need to somehow keep track of and manage data, much of which will continue to live in relational stores. As much as there is grown for key-value, graph, streaming, document, and other styles of data stores, we still use relational databases for many tasks. I don't see that going away anytime soon, especially for legacy software that will continue to live for some time. There is also the aspect of a DBA that involves the cleaning, checking, and verifying that data is in the form and format we expect, not to mention the need to ensure the data is intact and protected, and not changed inappropriately (or corrupt). Every system needs people that can dig through data and help decode its meaning as well as ensuring that business users can get it in to a report of some sort. Those tasks don't go away. While end users might have more self-service query tools, that doesn't ensure that they will actually know where the data is or how to reformat it. No end user tool is going to fix mm/dd/yyyy for a user expecting dd/mm/yyyy. There is also the need to just manage large amounts of data. I know data scientists have struggled to work with large scales of data in the past, and not just because of hardware limitations. They spend an inordinate amount of time gathering and organizing data that is used for analysis or AI/ML operations. Future organizations won't want to waste their time on those tasks and turn to some sort of DBA to help manage the large data needs of those people performing high level analysis. The DBA isn't going away, though the job is changing. Checking logs and backups is likely something you'll do less of in the future if you work in Operations. Instead, you'll manage more data, and you'll certainly need some coding skills to ensure you can perform in GitOps environments, where we don't click on things, but rather use code to define our systems and the configuration that we expect. The evolution can be daunting, but it can also be exciting. You have time to learn and change and build new skills. Embrace the opportunity to grow and change in your career as a DBA. Steve Jones - SSC Editor Join the debate, and respond to today's editorial on the forums |
