Home Uncategorized T-SQL Tuesday #200 (v1.0): Last of the DBAs

    T-SQL Tuesday #200 (v1.0): Last of the DBAs

    26898
    2

    tsqltuesday8:22 a.m. It was a dark and stormy night.

    Well, not so much, though Mike’s excruciating hangover was making him wish otherwise. The early morning sun shone brilliantly in through the office window, punctuating the headache that even four ibuprofen pills simply refused to touch. Sure, that hangover “miracle cure” medicine had come out last year, and everyone seemed to love it, but for Mike newer wasn’t always better.

    After taking a cautious sip of water, he carefully typed his password into the Windows 11 login screen, watched as his desktop sprang to life, and got ready to start his day.

    One thing Mike’s hangover had going for it? The office was quiet. Like, really quiet. Mike was the only one who’d bothered to trek into the physical space this Tuesday morning. He thought fondly about the earlier, and somewhat more social, days of the firm. 2008, when he was hired as the first DBA. 2013, when they used to regularly pull all-night rotations while migrating all of the databases into Availability Groups. Even as recently as 2020, his team still regularly got together for cloud migration and support planning meetings. These days, though, that just didn’t happen. Fuel prices had shot through the roof after the crisis of 2022, and Full Reality headsets made virtual meetings feel almost just like they used to, you know, in the real world.

    The only thing missing was the donuts. Mike really missed the donuts.

    Logging in to SQL Operations Studio, Mike checked his few remaining Agent jobs and was relieved to see that no failures had occurred the evening before. The firm’s final remaining Kimball Method data warehouse wasn’t used very often, but still ran a few key quarterly reports. Mike, as Lead DBA, was responsible for keeping it up and running, and for fixing the occasional blip.

    pexels-photo-48566Sitting back in his chair, Mike decided to close his eyes just for a bit. Why not? There wasn’t much left to do. He vaguely remembered being stressed some days, earlier in his career, but that all seemed like a dream now. He used to like that stress. That feeling of being needed, being important. But these days his once-thriving team of nine DBAs – five in the home office and four in other parts of the world – was not such a team at all. Mike was still Lead DBA by title, but he had only himself to lead.

    The beginning of the end for Mike’s DBA group was some time around 2019, when the executive board made a wholesale decision to start moving everything into what was then referred to as the “public cloud.” Some of his direct reports submitted their resignations letters in the following weeks. A few held on for a couple of years, helping with the migration work. But in the end, most had gone elsewhere in the firm, and were working closer to the business than ever—and, Mike figured, probably making a lot more money.

    Kelly, one of Mike’s sharpest Senior DBAs back in the late 2010s, was the biggest success story. She had really embraced all of the new technical challenges, showing up at the office with Hadoop and MongoDB books as early as 2016. As soon as the cloud move was announced, she volunteered to spearhead it, and management had taken her up on the offer. She was now in charge of the company’s small team of Data Engineers: six people who, Mike had to admit, were getting a lot more work done, a lot faster, than his nine ever could have dreamed of. They didn’t have to deal with storage devices, network switches, and all of the other hardware challenges Mike used to love to play with. They developed Data Flow and Visualization packages, helped the business implement a solid Data Studies practice, and a couple of the Engineers had created some very cool learning algorithms that produced results faster and more cleanly than any of the firm’s competitors. Well, so Mike had heard. He didn’t really understand a thing they did.

    9:43 a.m. Still a bit too early for even a really early lunch. Hoping to find something interesting to do with the rest of the morning, Mike poked around some DMVs. Maybe he’d find an index to optimize or something? A few end-user queries had run overnight, and one had even hit a suboptimal plan. But Query Adaptation Services, the killer feature that had had everyone really excited in 2020, had done its job. Glancing at the new execution plan, Mike couldn’t help but envy the artificial intelligence that Microsoft had baked into the Adaptation engine. There was no way he ever would have thought to hint the query to get a plan like that.

    10:32 a.m. No emails since yesterday. Nothing left to check. Mike vaguely wondered if he should browse around, maybe try to find a book on Python 4? Too tiring to think about starting over. Might as well go for a walk and take a really long lunch break…

    It’s 2026. The DBA is dead, dead, dead.

    Yes, that’s right. The DBA is gone. And good riddance to the DBA.

    That doesn’t mean that you are dead. That doesn’t mean that your career is dead. Quite the opposite. Data is going nowhere.

    You are a data professional. A data engineer. A data specialist. You are smart, creative, and agile. You provide business value. You are able to adapt to changing times and changing tides.

    The term DBA is a relic, a leftover from another age. An age of expensive infrastructure, high latency networks, and server time. Lots of server time on lots of servers. Servers that had to be administrated.

    DBA represents slow-moving, rigid IT policies that aren’t in line with the immediate needs of the business. DBA represents everything that used to be true of data, and nothing that still is.

    It’s 2018. Are you still a DBA?

    This month’s T-SQL Tuesday challenge, hosted by me, was to envision the world 100 months in the future.

    I see a future very bright for data professionals. We’ve seen an upward trend in data volumes, and that’s not stopping. We’ve seen an upward trend in requests by the business for more insights, faster. That’s not going anywhere. We’ve seen a move towards newer, faster, better, more self-driving technologies. And most importantly, we’ve seen a move away from traditional infrastructure and infrastructure roles. Databases of the future will not need “administration.” They’ll need many, many other tasks. In many cases, there won’t even be a “database” in the way we think of them today. And these are all good things.

    Being a data professional doesn’t even remotely mean that you have to be a “DBA.” It never did, yet somehow the two have become conflated. DBA friends, I want you all to wildly succeed. I want you to succeed with me—in our DBA-less world of the future. The DBA is dead. Long live the Data Professional.

    See you in 2026.

    Previous articleInvitation: T-SQL Tuesday #100 – Looking Forward 100 Months
    Next articleLooking Forward 100 Months (T-SQL Tuesday #100): The Roundup
    Adam Machanic helps companies get the most out of their SQL Server databases. He creates solid architectural foundations for high performance databases and is author of the award-winning SQL Server monitoring stored procedure, sp_WhoIsActive. Adam has contributed to numerous books on SQL Server development. A long-time Microsoft MVP for SQL Server, he speaks and trains at IT conferences across North America and Europe.

    2 COMMENTS

    1. Traditional DBA positions will diminish as SaaS adoption grows. Interesting perspective,great post.

    Comments are closed.