Voice of the DBA

Category: Editorial

  • AI Sings the Blues

    Quite a few of you aren’t thrilled or enamored by GenAI (Generative Artificial Intelligence) with things like ChatGPT or CoPilot. Some of you love it or find it fascinating. My guess is many of you are like me and you’re not quite sure how useful this is with a healthy dose of skepticism.

    No matter how you feel, I find this piece on AI Blues fascinating to read. It talks about some of the problems with AI and how people are becoming less tolerant of the small errors that AIs make. I think people might be upset with some of the issues, but I also think these are types of things we might see in a colleague as well, and we might tolerate them for much longer than we think.

    The thing in the piece that really caught my eye is that training models to make less errors might make them less creative. I don’t think they are very creative, but I do think part of the idea is that they might do something better than we could do, at least for those of us working in a space where we’re a beginner to intermediate (or advanced beginner). In many cases, we’d expect better output from a model.

    The note about training not being often or in-depth enough as well is interesting. Perhaps some of the models ought to be trained separately, with some having a more specific focus, like learning Python and producing very few errors, or learning Python to produce interesting solutions to large scale problems even if there are some errors in the code.

    I could see similarly more tightly trained models for other areas, such as marketing email or legal issues. Maybe we need a lot of different models tailored for different jobs. Of course, if companies are struggling to generate revenue on AI already (and I think they are), then can they afford to train and retrain specific models focused for smaller audiences?

    Maybe, and maybe Moore’s law will help bail out some AI experiments, but I’m not sure. For now, I continue to experiment, and I think skilled tech people ought to do the same. I don’t think it will replace us, but I do think it can be a useful tool in many ways to help us become more productive.

    Steve Jones

  • A Simple Solution

    In general, I like simple solutions to problems (and simple songs). They can take work, but a simple solution means it’s easy to explain to others, easy for others to refactor/modify, and we limit the amount of effort required to ensure others can use the solution. Sometimes I see very clever engineering solutions, but they become cumbersome to implement and maintain, or worse, no one other than the author understands how they work.

    In any size organization, that’s a piece of technical debt that can cause problems over time.

    When I see a simple solution, I feel a bit of joy. I got that recently with a blog post from Microsoft on how to transfer your CLR code to Azure SQL Managed Instance. While moving CLR code can be a pain, as you are moving a binary object, this post suggested using MI Link. This is a feature designed to replicate between a local SQL Server and MI, which isn’t something I think a lot of people want, but some do. I don’t know if I’d want this permanently set up, but as a way to move code around, this is very simple and easy way to move this code in a one-time shot. Perhaps as a way to set up a test from a local instance to the cloud.

    This isn’t my ideal way to move code around, as I prefer to have a specific pipeline dedicated to deploying code. However, I also think that this is a nice simple solution that might be easier in some situations. I like ease, and I like effectiveness, as long as I’m not sacrificing something more important, like performance.

    I see lots of creative solutions built by customers, often built by some smart people. However, they become quickly complex, and they are hard to explain to others in the organization. It can be difficult to trust anyone other than the author(s) to maintain the solution across time. It can be even more challenging when new staff onboard and never get an explanation of how or why something works.

    I aim for simple solutions, and those that I can easily explain, and others can easily pick up and understand. I’m curious today if you’ve seen complex solutions that were overkill, or ingeniously simple ones you admired.

    Steve Jones

    Listen to the podcast at Libsyn, Spotify, or iTunes.

    Note, podcasts are only available for a limited time online.

  • Casino Night is Back

    We are bringing back the SQL Server Central Opening Night party at the PASS Data Community Summit this year. This was one of the most fun things I’ve been a part of at the Summit for many years and I’m looking forward to the conference even more.

    The party will take place Tuesday night, after the Welcome reception. from 8pm-10pm, once again catered as a casino themed party. I’m using the same company we’ve used in the past, and we’ll have a cash bar and lots of prizes. I’ve already started to make a shopping list of fun things to give out as random prizes as you enjoy the games. In the spirit of competition, we will save a prize or two for the top chip winners.

    We will have a limit to the number of people that we can host, as this is a community event, and I do have a budget. Rather than using the referral codes as we did in the past, this year we’ll have a lottery. You’ll have to be registered for the PASS Data Community Summit, but if you are, and want to come, then you’ll need to fill out an entry form for the lottery.

    We’ll close the lottery on 21 Oct and then verify all the entrants are registered before randomly picking 150 of them. We are also working on a sponsor (for more prizes!!!!) and they’ll have a few tickets to give away as well. If you want a Golden Ticket, you’ll need to enter and cross your fingers. We only have limited space and may have a Scary DBA checking tickets at the door.

    If you’re on the fence about attending the Data Community Summit, maybe this will convince you to push on your management to send you. This is a huge data conference where you have the chance to learn, connect with others, and return to your job energized with lots of ideas, solutions, and inspiration for the future.

    Good luck to everyone and I look forward to seeing the lucky winners at the Casino Night party.

    Steve Jones

    Listen to the podcast at Libsyn, Spotify, or iTunes.

    Note, podcasts are only available for a limited time online.

  • The Invisible Disk

    I saw a post recently about the incredible shrinking computer, where a software consultant had written down his predictions for the future. In this case, he writes about a computer shrinking smaller than today, essentially to one chip, with docking stations wherever you need them. We’re already at the place where many of us have an extremely powerful computer in our pockets in the form of a mobile phone. I’m still amazed that I bought a phone recently with half a terabyte of storage.

    A few years ago I gave a keynote where I looked at the changes in disk storage, starting with an IBM hard drive being loaded onto a plane. The capacity was 5MB. I remember working with KB floppy disks and MB hard drives, and lots of different connection technologies: IDE, ESDI, SCSI, and more. I’ve seen disks shrink from a 5 1/4″ form factor to mere wafers with today’s NVMe drives.

    However, what might be even more interesting is that I don’t even think about drives anymore as being a physical thing. Certainly, my mobile, my laptop, and my desktop have physical drives, but they’re all connected together, essentially giving me invisible disks. I can take a picture on my phone and it shows up on my laptop or desktop in seconds. The disk might as well be invisible.

    As I left full-time Operations work, we were starting to see this in our servers, which were essentially invisible VMs stored in some location, on storage that we never laid eyes on. We never saw disks fail because the storage arrays had redundant disks. There were people replacing failed drives every week to ensure that service wasn’t interrupted.

    These days we rent storage in the cloud without ever working with the hardware. Most of us who specify the configuration in the cloud don’t think about disks. We think about the total capacity and the IOPs that we need for our system. We care about the results, not the hardware.

    Most of us. Some of us care and worry about it, but a lot of the people I know in data work, even as operational DBAs, don’t worry about hardware. They need a high level of service and low latency to ensure clients can retrieve data. Even when requesting more hardware, most of us are asking for more capacity or throughput, not specific models of hardware.

    I find myself rarely caring about disks anymore, though recently I was glad I had a flash drive in my bag. I needed to transfer a file to some air-gapped computers, and a quick copy on an old 2019 PASS Summit recording flash drive worked well. That drive was certainly visible, and full of memories.

    Steve Jones

    Listen to the podcast at Libsyn, Spotify, or iTunes.

    Note, podcasts are only available for a limited time online.