Voice of the DBA

Category: Editorial

  • The Black Boxes

    Machine learning and artificial intelligence seem to be the hot topics these days. From bots that can interact with people to systems that learn and grow as they process more data, it seems that science fiction is becoming reality. At least, in limited ways. Autonomous cars, perhaps the highest profile example of these topics, are advancing and being tested in a few locations around the world, but I think we are a long way from having human controlled and autonomous cars interacting freely at any scale. There are still plenty of issues to work out, and the consequences from mistakes require serious consideration.

    I was thinking of this as I read an interesting question: Whose black box do you trust? It’s a look at algorithms and machine learning, and the impact they have on the world around us, despite many of us not understanding how they work. The main examples in the piece are in the area of journalism as it relates to social media (primarily Google and Facebook), but also touches on autonomous vehicles, both autos and planes. The latter was a bit of a shock to me as I assumed humans always handled takeoff and landing, something the author says doesn’t happen at SFO. Some searches around pilot sites seem to note that automated landing is done regularly to test systems, but is used in a minority of cases.

    The question is, do we trust the black boxes that run our systems, and really, does it matter? In the piece, Tim O’Reilly says he has four tests for trusting an algorithm:

    • the outcome is clear
    • success is measurable
    • the goals of the creators align with the goals of the consumer
    • does the algorithm lead creators and users to better long term decisions.

    Those are interesting ways to evaluate a system, though I think the problem is that the last two are a bit nebulous. One of the things that I see more and more as I get older is that the same data or the same facts can lead two different people (or groups) to two different results. Our goals, our interpretation of events, even the weights we place on the various factors in a complex system vary dramatically from person to person. In such a world, can we truly evaluate what the goals of a creator are? Forget about consumers, assume one person building a software system. They will have multiple goals, and do we really think their goals can be easily listed? Or weighted/ranked appropriately? What about when the goals change?

    I really think that the black boxes need more disclosure, though I freely admit there isn’t a good way I know of to do this. However, I do know one thing that can be better disclosed: data. We can have more openness and analysis of data from software systems along with some accountability by creators for the impacts of their software. Again, I don’t know how to enforce accountability, especially at scales that encompass millions of consumers and easily cross country borders. That is a problem I think we need to find ways to tackle, at least at some manageable level. Maybe using the 80/20 rule where 80% of consumers and creators find the outcome to be a good one.

    The world of technology and software are advancing and growing extremely quickly. Certainly hardware advances, but it seems the last 5-10 years have been more about new and different software applications that fundamentally alter the way humans can interact in social, business, and government situations. Underpinning all the changes is data. New data, more data, and novel ways of working with this data in ways that were unheard of 20 years ago. It’s an amazing time to be a data professional.

    Steve Jones

    The Voice of the DBA Podcast

    Listen to the MP3 Audio ( 5.2MB) podcast or subscribe to the feed at iTunes and Libsyn .

  • What Helps You Learn?

    There was a post recently asking how to train a junior DBA. What tasks do you give them, how to get them productive, etc. The responses were interesting, but I was struck by one note from a person that is a junior DBA. This person mentioned a few things that have helped them get better at their job, some of which I hadn’t thought about before.

    Since I try to teach people about SQL Server in writing, in presentations, and longer formats, I decided this might make a good poll question. If you spend a few minutes thinking about how you’ve improved your skills at your job, let us know what works best for you.

    What things can others do to help you learn more about your job?

    Are there specific things that have actually worked better for you? Perhaps you prefer to learn alone, but even then, what works well for you that builds skills. Is there a format or style of presentation that reaches you?

    I’m a big believer in multiple voices and multiple ways to teach people. This is one reason we sometimes have 2-3 articles on the same topic from different authors. Each will reach different people, and speak to them in different ways. I think the format (text, video, live, etc. ) also has some impact on what helps different people learn. Let us know today what works best for you.

    Steve Jones

    The Voice of the DBA Podcast

    Listen to the MP3 Audio ( 2.1MB) podcast or subscribe to the feed at iTunes and Mevio.

  • Capture the Flag

    I’ve read a lot of science fiction in my life. Quite a few books talk about cyberspace and the interactions of humans and software. In Snowcrash there are AI systems defending systems. The Unincorporated Man series has AI systems that not only have sentience, they hide their capabilities from humans. Numerous other examples exist, including movies such as The Matrix, which popularized system programs independently able to evaluate and react to security issues.

    Science fiction is here, showcased recently at the DARPA Cyber Grand Challenge. You can watch a bit about the challenge or read a short article about what happened. Overall, it’s amazing. Seven systems ran various pieces of software that were targeted at finding vulnerabilities, patching their own and exploiting those of other systems. All without human intervention.

    In a visualization, I picture a team of Agent Smiths (from the Matrix) seeking out rogue programs that seek to wreck havoc inside of the operating environment. Or maybe a series of automated developers tracking down bugs, such as buffer overflows and correcting the code to remove the issue.

    Is this where static code analysis is heading? Automated repair, or maybe even real time repair? Could we actually have a software bot that might run through your Entity Framework (EF) application, tracking down SQL Injection vulnerabilities and correcting them? Even at a rudimentary level, I can imagine a bot that examines incorrectly parameterized queries and rewrites the code to properly manage the .NET code. Perhaps at some point we’ll even have bots that can understand queries from some application and build stored procedures on the fly that accept parameters, are called the next time a particular method is called, and can even grow to evolve with schema changes.

    I know the simplistic ways in which we might build this code would have their own sorts of issues, but I’m hopeful that at some point we’ll have semi-intelligent bots that can fix code to ensure it’s secure and reliable, while keeping the creativity originally built by the human developer.

    Steve Jones

    The Voice of the DBA Podcast

    Listen to the MP3 Audio ( 3.2MB) podcast or subscribe to the feed at iTunes and Mevio .

  • Analyzing Data in Real Time

    A few years ago I was listening to a presentation from Microsoft where they touted a customer that had real time analytics with Analysis Services. Since my exposure had been that cubes needed to be processed, a resource intensive process, I was wondering what this meant. Imagine my surprise to hear that the cube was being updated within 5 seconds of the changes in the OLTP system. Talk about real time.

    Of course, the definition of real time isn’t really well known. We linked a piece in this week’s newsletter that looks at a survey of companies and their investment in real time, streaming, data movement pipelines. The definitions from companies about real time range from sub-second to 24 hours. I’m sure I think that 24 hours is real time, but if you’re used to processing cubes weekly, maybe that feels real time. I tend to see real time on the order of minutes, maybe a few tens of minutes. Once we get to hours, that feels more like batch processing, albeit fairly rapid batch work.

    I am sure there are systems that need real time analysis, especially in some areas where decisions must be made quickly. As we move to more and more machine learning algorithms and automated intelligence, there will certainly be more call for real-time data movement. Yet another opportunity for data professionals, especially those that work with ETL pipelines. There will be tough problems, not only with moving data, but tracking lineage, recovering from issues, even weeding out bad data quickly.

    However, that’s an opportunity for you, not me. I bet there are a few data professionals out there that feel the way I do. There are great challenges in solving real time problems, and building systems that can handle high volumes of data. The thing is, I value my sleep. I value not working all-nighters, I value a balance in my life. While I find the problems fascinating and the money involved tempting, I prefer to work on more pedestrian systems.

    I’m glad there are people that want to work on very difficult problems, and I wish you all the best in taking advantage of these opportunities. I hope you’re well paid, and you have a great time building these impressive systems. I also hope to read about some of the amazing things you do, so please, share the knowledge where you can. This is a great, exciting time to work with data, and it’s one that I continue to enjoy every day.

    Steve Jones