Because I’m All About That (Data)base.... — SolarWinds TechPod 074

Stream on:
Is your database a junk drawer? Is it more like a black hole? If you’re afraid to look in there, join TechPod hosts Ashley and Sean as they demystify Database with the help of Head Geek and resident Database expert Thomas LaRock. This episode will discuss the challenges database administrators face, and the tools they can use to address them.  © 2023 SolarWinds Worldwide, LLC. All rights reserved  RELATED LINKS:
Thomas LaRock

Guest

Thomas LaRock is a Head Geek™ at SolarWinds and a Microsoft® Certified Master, Microsoft Data Platform MVP, VMware® vExpert, and former Microsoft Certified Trainer. He has over… Read More

Host

We’re Geekbuilt.® Developed by network and systems engineers who know what it takes to manage today's dynamic IT environments, SolarWinds has a deep connection to… Read More
Sean Sebring

Host

Some people call him Mr. ITIL - actually, nobody calls him that - But everyone who works with Sean knows how crazy he is about… Read More

Episode Transcript

Audio: This episode of TechPod is brought to you by the SolarWinds THWACK community. Join us at our free SolarWinds User Group in London on April 19th and 20th at the Sheraton Skyline Hotel London Heathrow. The event will feature two days of sessions, including keynotes from SolarWinds executives, tech geeks, and other product experts to answer all your questions about your SolarWinds solutions. To find out more, go to thwack.solarwinds.com and click on the Events tab.

Sean Sebring: Hello and welcome to another episode of SolarWinds TechPod. I’m your host, Sean Sebring, and I’d like to reintroduce Ashley Adams, who was with us last episode as my new partner in crime and co-host.

Ashley Adams: Thank you, Sean. I am very honored to be co-hosting the SolarWinds TechPod with you. I know I have some big shoes to fill with Chris’s departure, but I’m very happy to be a part of keeping the SolarWinds TechPod machine going. If you listen to the last episode, you probably got an idea of who I am, but for any new listeners, I am a staff product marketing manager here at SolarWinds, joined almost three years ago to the day, and I have about 12 years of experience working in all things IT, both out of here in Austin, Texas, as well as in Paris, France.
We’re super excited to kick this episode off. We have with us Thomas LaRock. He is a head geek here at SolarWinds. If you’re familiar with our THWACK community, lots Orange Matter blogs. He’s written among other things. He is a Microsoft Certified Master, Microsoft Data Platform MVP, VMware vExpert, and former Microsoft Certified Trainer. He has over 20 years experience in the IT industry as a programmer, developer, analyst, and database administrator. Tom, thank you for joining us and welcome to the show.

Thomas LaRock: Thanks for having me. You make me sound old, though.

Ashley Adams: Lots of experience always makes people sound old, but that doesn’t always have to be the case.

Sean Sebring: Wisened would be another way to phrase it.

Ashley Adams: There you go.

Thomas LaRock: All right, that’s good.

Ashley Adams: I guess today we’re going to attempt to unmask the infamous database and we’ll be discussing all things, including tools, personas, how database fits into the bigger picture. Would you like to start us off? What is your data origin story? At what point did you feel destined to live a life of database?

Thomas LaRock: Nobody really ever wants to become a data janitor like I am. Nobody sets out to do that. You just all sort of fall into it. For me, my data origin story really just goes back to how, even at the earliest ages, I can just remember being… Math came easy. Numbers and math sort of came easy. As you progress through school, that becomes higher level things like statistics and calculus. I ended up going to college to major in mathematics, and the biggest reason to do that was, because in math there is no writing. It’s just, “Here you go. Two plus two is four. That’s it. I’m done.” These other majors, you had to do term papers and all sorts of stuff and, no, not me. Here you go, here’s a formula, here’s the answer, I’m done. Out.
Then after I left grad school with my master’s in math, I got a job as a programmer analyst with this consulting firm. I was hired by a guy. I had no real practical experience. I’m not computer science major, but this guy was a physics professor at one point, and he looked at me and said, “Oh, you’ve done some math. You’ve worked with an astronomy.” He goes, “I can teach you whatever I need, so you’re hired.” And that was great, so he taught me some coding. My first system was… I was working PowerBuilder on top of Sybase. I’m sorry, the artist formerly known as Sybase. That’s where I really started to work with databases in general.
From there was PowerBuilder in Oracle, then back to some PowerBuilder in Sybase, then into SQL Server. Around the time I was working with SQL Server, I was unofficially a junior DBA, and then I became the real DBA because the other DBAs quit. By the way, when all the DBAs in the company quit, you should consider that a red flag. You should ask questions. I didn’t ask any questions, I said, “Sure, I’ll be the DBA. You guys just reset passwords and surf the web all day. Sounds great.” I got tired of being the programmer because you had to keep learning new languages all the time and platforms. I’ll just go work with these databases. Nothing ever changes there, right?
So, that’s how I fell into becoming what essentially I am now. I just say I’m a data janitor. I moved data around from one place to another, I clean it up, I pick up after other people’s mistakes, things of that nature. That’s how I got to where I am today.

Ashley Adams: I love how you landed a promotion by default. That happens sometimes.

Thomas LaRock: Yeah. Seriously, I was the only one in the company that had done a restore, a backup, and reset passwords. That was it. They were like, “Done. You can recover a database. Congratulations.” That’s how most DBAs get started is they’ve done one of those three things and the company’s in the bind. “Great. You’re hired.”

Sean Sebring: Well, I guess I can work on my resume. I haven’t done any of those things, but I’ll note that I can put one of those in to qualifying. I did want to ask because that’s awesome. To get us started, what is a database? Am I saying it right?

Thomas LaRock: You’re saying it perfectly. When I hear that, I think I can charge three or four times by normal rate.

Sean Sebring: A database. Okay. Okay.

Thomas LaRock: The more words I hear like that, the more I realize somebody has no idea about any of this, and whatever I say, they’ll just believe.

Sean Sebring: Well, we’re going to have perfect synchronicity here then. I would ask, can you explain it to me like I’m five? What is it?

Thomas LaRock: Oh, what is a database? Think of it this way. Really, there are different types of databases. Let’s just think of it as database as a place to store your data. That’s all. That’s all it is. Data could be anything. It could be songs on your phone, it could be pictures on your phone, it could be ones and zeros. It could be just about any piece of information you can imagine could be stored in the database of one type or another. That’s it.

Sean Sebring: Okay. I guess a good follow up, Mr. Data Janitor is, why is data and the databases such a pain in the behind?

Thomas LaRock: Yeah. Data is hard.

Sean Sebring: Well, I mean, the way you described that, it sounded fairly simple. It’s got to be more complicated than just, “Here’s the container with all my photos in it.”

Thomas LaRock: Yeah. Here’s a container with all your photos in it, which ones are selfies? How would you know? Would you have to look at each one? You got a couple thousand photos and you all of a sudden say, “Hey, I remember a selfie of me… It’s got to be in here somewhere.” How would you go about finding it?

Sean Sebring: I imagine that querying is a big part of why you need clean… And janitors is a good word for this, I’m seeing now. Clean database. I imagine it’s also not just a human that’s going to be querying this database. It may very well be a system trying to query the database or another application.

Thomas LaRock: Then it gets messy from there, because I gave a simple example of how we identify a selfie, and the easy answer is it’s going to get tagged that way. Now, are you going to tag it or maybe you could have a computer tag it for you, and then you could just search by tags? That sounds great, but maybe what you think is a selfie, somebody else doesn’t think is a selfie. Now you might miss that piece of information and now start thinking about things of, “Okay, here’s a piece of data. I think that is a piece of personal information and it should be secured and kept private.” And somebody else goes, “Nah, that’s just some customer data. Don’t worry about it.” Well, these are two wildly different things, so now you’ve got people having to discuss what this piece of data means.
An example I give from my own experience is I worked financial services for many years as production DBA, and every so often somebody would have the great idea that we would need a data warehouse. We need one big data warehouse for the entire company. Sure we do, but what does that look like? Okay. They’d spend time talking about certain aspects of it, what this is, and that what this will look, all the different people use it. Then I could sit there and say, “Okay, great. We’ve got 13 different teams here. Define what cash means to you.” And one person real quick, “Oh, cash means this,” and somebody else “Oh, don’t forget about the air corrections.” “Well, are you talking next day or are you talking now? What about the sweep interest? Would that be a fact?”
You got all these different departments to which cash means money in my wallet if I have it, right? That’s cash. No, cash has many different meanings of financials industry. So how do you get all these people to agree upon what these definitions are and classify and categorize their data? It can get messy real quick.

Ashley Adams: That’s interesting. I was working in Europe when GDPR was first introduced and all of the regulations and database cleanup that had to go into deploying that for the first time. I can only imagine.

Thomas LaRock: Data security and privacy was such a second class citizen all the time for many, many, many years. Even with the European regulations before GDR, most of that stuff was just ignored like, “Yeah. Well, we aren’t really a business in Europe. We don’t have to adhere to any of that until you did.” Thankfully, many companies, including SolarWinds, have plenty of notice in order to become compliant, but a lot of people just have ignorance about it. Even to this day, they just don’t even care. It happens. Or they calculate and say, “What’s that first find going to cost us? It’s fine. Don’t worry about it.” Meanwhile, data leaks, data security and privacy leaks, they continue to happen at an alarming rate. Until you start threatening people with real punishment for this, then it’s just going to keep happening.

Sean Sebring: So the security and compliance, I guess, raises some challenges. As you’re kind of just mentioning, they’re different today. We actually have some standards in place today, so they’re different today than they were 20-ish years ago. Where do you see the biggest challenges as far as keeping that secure? You also mentioned a data warehouse previously, so I’m guessing there’s different ways to store things in databases. Maybe that’ll be another contributing piece to this.

Thomas LaRock: You’ve asked a few things there. Let me see if I can address them all. Yes, there’s many different ways to store your data. Databases typically deal with some type of data that might have structure. Typically, a relational database means that there’s a relation between the entities identified in a database, what we usually call tables. And table is nothing more than a way for a human to understand how the data is on disk. It’s just a logical representation of data on disk. That’s what a table is. And tables have relations, that’s why we call it relational databases. But not all databases are relational. Some of them are what we call flavors of NoSQL, right? A document store isn’t necessarily a relational graph database, isn’t necessarily relational.
You could also have just unstructured data in general, like images. There’s no real structure to an image. You could argue there is, but there isn’t. Your JSON or XML, these are typically things called unstructured. There’s a lot of different types of data out there. The real question is, how do you go about identifying and classifying what might be sensitive or confidential or GDPR or PII? Because there’s so many different rules and regulations around all that. The biggest problem with all that, I might be able to tell you. I can go into a database and say, “I see what we call social security numbers.” In other countries, they’re called government ID numbers or whatever. But let’s just say that’s a piece of information that you might want to keep private.
More importantly, you’ve got to communicate to somebody else that this is a piece of information that’s confidential or sensitive and this should be secured. The biggest problem with that until recently would be the tools that would allow you to do this. A lot of companies, specifically Microsoft, are making a better effort at helping you to classify and categorize this data. Until we get to that point where it gets a little bit easier for people to have that role, it’s always going to be a struggle, I feel, for people to achieve that level of security and privacy that’s necessary for their data.

Ashley Adams: I’m curious to what level you think people in general care anymore, because I feel like even among my friends, we do everything online now. I just recite my insurance policy, which has all of my data all online with e-signatures and data’s going everywhere. Do you think that there’s generational gaps in people who are more concerned or less concerned, or even companies making it evident to people that they should be concerned about where their data is going on a daily basis?

Thomas LaRock: Oh, there’s definitely generational gaps. My parents are in their 80s. They have no concept of the information that is being shared or how it’s getting used. People younger, like my kids, they’re just used to it. They’re like, “Yeah, this is what happens. I’m watching YouTube and this, that, and the other, and of course I’m going to get these ads for this other thing.” They understand why this stuff is happening at least, but I’m not sure that they always care to stop it. Sometimes they do, sometimes they’re like, “Yeah, no, I don’t want that. No, I’m not clicking on that.” They’re getting a little more aware as they get older.
For the rest of us in, we’ll just say our experienced stages of life, we might be more in tune with it. Like you said, there’s a lot more stuff online. You’re sharing a lot more, but there’s also an element of trust. “I really trust that this company is going to do something to protect it.” Me personally, if I don’t trust that, I ask questions. I am… What’s that, AITA? I am TA in that scenario, right? I will step up and just say, “Tell me more about how or what you’re going to be doing with this data, or what happens after it.” There are a lot of places where I just don’t even trust this, so I don’t even bother to engage. I just say, “Nope, not interested in letting you have it.”
I think where I get most offended is when I’m added to a marketing email list. When I’m opted in by default, that just bothers me. I should not be forced to take time for my day to click buttons and go to websites and add an email and say, “Unsubscribe,” and then get that thing. “Oh, well, an email could have already been in the queue. It’s going to take 10 days.” “10 days? I can help you with that. I work with data all the time. I can remove my stuff from your system now if you’ll let me.” It just so annoyed by… That’s probably what really riles me up.

Sean Sebring: I can sense it. I can sense it, Tom, the angst, the frustration. Also, I sympathize. Yeah.

Ashley Adams: Oh, well, I was joking with a friend. The messages they send you now, too, when you hit unsubscribe, they’ll say things like, “Oh, you don’t want discounts for this?” It’s very threatening language that comes with the unsubscribe button, which I don’t need in my day either.

Sean Sebring: Yeah. A little passive-aggressive in there.

Thomas LaRock: Yeah. But they word it in a way to be confusing, which, if it’s not, it should be against the law. It should be very clear what you’re clicking and accepting or agreeing or telling them not to send you.

Sean Sebring: With databases, I get the need for the security, the privacy, and the… I’ll call it the hygiene. I don’t know if that’s a good term for your reference of janitor. It makes sense. The hygiene of the data, I think I’m getting a better understanding of that. What kind of technology challenges from supporting a database do you have? Not just from good data, but what kind of challenges does the technology give us?

Thomas LaRock: Challenges around data security and privacy specifically, or just challenges around data in general?

Sean Sebring: I guess it could be around both. All of the above. I’m just curious, as a data janitor, you’ve walked the halls, so to speak, for so long. What challenges do you have there?

Thomas LaRock: Here’s a good example. In SQL Server, Microsoft SQL Server, there is a data type called timestamp. Makes sense, right? If I say timestamp, you know what I mean, right?

Sean Sebring: Yeah. Yeah. I know what a timestamp is. I’m a timestamp pro.

Thomas LaRock: Okay, cool. Oracle would also have a timestamp data type, right? Now you would think to yourself, “Okay, so if I have data in Oracle and that’s my source system, and I’m going to take that data, I’m going to extract it, I have to do some work to it, and then I’m going to store it inside of SQL Server,” you would think to yourself, “Not a problem. Timestamp to timestamp.” No issues, right?

Ashley Adams: Seemingly so. Yeah.

Sean Sebring: I’m sensing the direction [inaudible 00:17:48].

Thomas LaRock: Yeah, exactly. You know what’s coming. Guess what? The SQL server timestamp data type is not what we would think would be like the day and time that something happened. It’s more, I think, it’s like a [inaudible 00:17:59] or something. It does not represent what you think it does. There are plenty of examples of where data types inside databases have the same name but aren’t really the same thing. But if you don’t have experience, decades of experience of working with these things, you don’t know that right upfront. So you spend your time, you’re working, you’re doing, so you go, “Why didn’t this work? What’s happening here?” And then you have to research and then you find things are taking a left turn.
Now, that’s just one example. There are so many examples of where assumptions are made and data gets mistreated accordingly. For example, this is the bane of my existence is a null value. I hate nulls. I hate nulls in everything they represent. I can’t stand working with null values. I will have people tell me how nulls are unnecessary, and I look at them and say, “No, they’re not. It takes work and we can get rid of them.” But null values cause a lot more problems than they ever solve. The problem isn’t just the storing of the data in the databases, but it’s the tools that move data from one to another through things of they make assumptions based upon this thing called ANSI standards. If I tell you that when you connect to SQL server, the default is ANSI_NULLS are off, do you know what ANSI_NULLS off means? And the answer is no. But if I tell you that you fire up Excel and the default is ANSI_NULLS are on, you still don’t know what that means, but at least you might think to yourself, “Hey, if they’re on in one tool but often another, is this possibly an issue?”
I’ve given examples. I actually don’t know what the defaults are, but this is what happens is people build tools and they go, “Whoa, it’s adhering to the ANSI standard,” but a different expectation for each tool, which then results in how data gets transferred. This goes back to where people where in… If you ever used Linux from Windows and you had to be storing stuff in binary and converting stuff before you could store it somewhere else, and you got to move data back and forth and do all that conversion, this isn’t a new problem. But this is why data janitors are necessary, because we have to come in and figure out, “Okay, what’s really happening here? How did it even happen? How can we minimize the opportunity from this happening again?” There are a lot of challenges just when it comes to data just for stuff like that. We spend a lot of time just cleaning up messes in the hall.

Ashley Adams: Sounds a lot like being proactive more than reactive on the database side, perhaps.

Thomas LaRock: It could sound that way until you get called at 3:00 AM because something didn’t work, to pick on a developer, but a developer pushed code to production, and now all of a sudden everything’s failing. What it was was that their code uses a Select Star, and the Select Star is expecting 10 columns, but they decided to add an 11th, and now everything downstream is failing. Clearly, it’s a problem with the database so you’ve got to come in and fix it, and you look at them and go, “So you made all these changes and screwed things up. What would you like me to fix? I mean, what am I supposed to do? Just re-architect everything for you here at three in the morning? I don’t know what the right thing is to do. I don’t know what the business requirements are. Could I solve your issue? Sure. Will it create 10 new issues? Probably. Maybe we should undo your change.” That’s the reactive part is where traditionally DBAs end up getting called when everything breaks. Now, these days in the world of DevOps, it’s more often the developer, the person who pushed the code to production, they’re more responsible for supporting it in those off hours. DBA isn’t as reactive these days. We do get to work on, say, more architectural or platform decisions, but we’re still called in for our expertise from time to time.

Sean Sebring: Something you had said was proactive versus reactive. Just based off of talking to you, Tom, and learning from what we’ve gone through today, it does make sense that database should be able to be more proactive than needing to rely on reactive actions. Just because, like you said, “I want to be part of the strategic planning. If you’re going to make changes, let’s prepare the database theoretically for the changes that you’re going to push so that 11th column can exist and be prepared to accept what you’re trying to send to it.” From a strategic standpoint, it definitely sounds like database should be part of those strategic decisions.
When we’re talking about observability, like the 3:00 AM call you’d hope to avoid, maybe from a standpoint of observing that full stack, we can pinpoint that it’s not the database’s fault that it’s not been designed to receive that 11th column of data. How can you talk to us about where database fits in the big picture of observability there?

Thomas LaRock: Sure. First of all, it’s never the database’s fault. Never.

Sean Sebring: [inaudible 00:22:57].

Thomas LaRock: You blame the network before the database, but that’s a different story. That’s a different podcast.

Sean Sebring: As long as the files are in the computer, I’m okay.

Thomas LaRock: No, the thing is the database does what you’ve told it to do. You’re the one in control of all of it. You’ve told it to do something. We had situations where the databases would just get blamed for everything. Now, I’ve got a few hundred databases. They’re all built the same, they’re all working the same, but there’s a problem with one of them, and you’re like, “Oh, there’s something wrong.” I know it’s wrong because the database is fine till your code and data was pushed to it. I know what the problem is, and the problem has to do with you and not with this actual database server sitting over here. Because that tiny database, that little service that’s running on a server, that thing’s just doing what you told it to do. If you don’t like what’s happening, you should think about what you asked it to do for you.
That’s the simple truth of it all. Now, is it possible that the database is misconfigured? Absolutely. In our case, I have 300-plus databases all configured the same. If something’s wrong for all of these, I should know that. But if you think one of them just needs to be configured differently, well, we should talk about that. The discussion shouldn’t just be, “Hey, I’m the developer, just make stuff happen.” That’s not really an answer as to why we should make a configuration change on the wider scale. That being said, don’t blame the database.
Now, with regards to me also mentioning how developers are being more on the frontline these days than historically they have been in traditional IT circles, how does the database itself fit inside of… Let’s just call this new world observability because that’s the latest thing. Systems, it’s DevOps and observability. That’s all you ever see. You end up at a AWS re:Invent, and that’s all anybody wants to talk about. “What do you do for DevOps? What’s your observability solution? If you’re talking DevOps, for example, if you haven’t started with DevOpsing your databases, so to speak, if you haven’t started at the database layer, I think you made a mistake. I don’t think you can be DevOps and be thinking to yourself continuous delivery, continuous integration. Oh, yeah, the problem with the database again. The database should have been the first thing front and center when you were building out that DevOps process inside your team. So, yes, definitely should be more of a first class citizen when those discussions are happening.

Sean Sebring: When you said it like that, Tom, and I’m just trying to put my five-year-old brain around the concepts here. Databases are containers, right? And if you’re trying to say, “I need to move this stuff in store this in a container,” if the container won’t hold it in the first place, then we have a problem. So you really need to start with having a appropriate container for the data. Is that fair?

Thomas LaRock: Absolutely. Right tool for the job. Right. The example I do give here would be graph databases. So SQL Server has the ability to do graph databases, but that doesn’t mean a SQL Server would necessarily be the correct choice for you to use if you have that need for a graph database. Perhaps Cosmos DB might be a little bit better or something else. You always want to use the right tool for the job. A lot of these database platforms these days can do a lot of different things, and that’s great, and it might meet all your needs, but every now and then you might come across an edge case and go, “I really need something slightly different.”

Sean Sebring: Okay. Continuing my analogies, I also like to talk about food as often as possible. If my mom made me a sandwich and I needed to take it to school in a container, a Ziploc could do it, but Ziplocs are squishing. It’s going to potentially mutilate my sandwich. A hard Tupperware could be a better container for my sandwich, which the sandwich clearly PB and J is the data we’re talking about here. That’s what we’re storing. But is that also fair? Am I getting that well enough?

Thomas LaRock: That’s fair. That’s absolutely fair. Although I would go with the Evel Knievel lunchbox, but that’s fine.

Sean Sebring: Oh. What if I put my Tupperware in my Evel Knievel lunchbox? How would that fit into database?

Thomas LaRock: Well, that’s like putting an Excel spreadsheet inside of a database, so that’s fine. That happens all the time.

Sean Sebring: Yes. All right. We’re making progress here.

Ashley Adams: It makes me think, too, about… when we’re talking about observability, digital transformation. Databases on-prem, in the cloud, moving from on-premises to the cloud. Can you talk a little bit about what that kind of means for the database world?

Thomas LaRock: Yeah, data is everywhere. Data and database is everywhere. The systems you have don’t necessarily have to be working with databases, but they probably are working with data. I mean, we had a system that pulled in 1,000 one-gig files every day. From there, it was pushed into an application where it was processed and analyzed and things of that nature, and that was ever before it even touched the database, before it ever got stored. So, you have systems that are using data from a variety of sources, storing it, moving it. It’s crazy, almost. It’s a lot of piping. It’s a lot of piping. That’s why we have the role of data engineers these days, people who specialize in the movement of data.
There’s always been a need to understand how data moves through a system through your entire company, from the customer endpoint, all the way to the data back again type thing, which commonly we talk to or say is full stack. Full stack can mean different things. To me, I think a full stack is all the different layers that all these applications and data move through all the time. Where I take observability to mean is simply having insight into all the different layers as data moves through.
For me, one of the hidden ones was always network. I worked inside of SQL Server. SQL Server has no idea what’s happening in the network; it only knows what’s happening inside its engine. It knows very little about disk activity. It knows when it’s trying to write or read from a disc, but it doesn’t know if that disk… Is it virtualized? Where’s that disc? It has no knowledge of that. It has no knowledge of network. It says, “Oh, yeah, I’m waiting on network.” Are you really? You don’t know. You’re guessing. So, that’s when you start needing these other tools to give you information about what’s really happening with storage network.
And then you need a view. To me, that’s what we call observability. Give me that view so I can see from beginning to end the entire picture for this process, this application. But as far as databases observability go, well, database is a black box. Nobody knows what’s happening inside of there. It’s very easy because you have no knowledge of it to sort of blame it and say, “Eh, I don’t know what’s happening here. I wrote this query. Things aren’t coming back. Must be a database problem.” Is it, though? When you have a dashboard that gives you that observability into all the systems, you can look and say, “No, it looks like you’re trying to query and the data now isn’t there anymore. It’s halfway around the world because we had the failover and the data hasn’t replicated yet.” It’s not really a database issue necessarily, but we had an event. This happened, the data’s going to get there, and we can work through what’s really going on as opposed to making somebody go figure out why one query is bad. If you have more insight, you’re more efficient at everything you do.

Sean Sebring: I love that. Yeah, that’s good. As IT professionals, really, what we’re here to do is support services, and services is that ephemeral thing that’s comprised of a ton of different things. I like the way you presented it from a perspective from the database. Everyone can kind of paint the same picture with different colors depending on their background. But you have so many supporting elements: the database itself, the network, even the power. The electricity being on is part of an equation to support a service. All of those together are what make the service. Again, the service itself is ephemeral. We call it a service, but you can’t just discover, you can’t just observe a service, unless you’re observing all of the components, these configuration items, we’ll call them, that make the service.
Database is, as I’m learning more and more, a foundational part of a service, especially now that there are so many things with integrations today. Ashley brought up digital transformation, so moving things from an on-premise local data warehouse to somewhere else, the need to move the stuff, being able to talk to other applications, other systems, database. Those are all foundational pieces. Everything might have a database and all these systems might refer to things differently. I call that field mapping in my simple brain. We both have the same stuff; we just call it different things and maybe you use the European date format and I use the American date format. All of those little things are what might need to be observed, so I guess a good next question kind of revolves around personas for me. What types of industries themselves… And I’m guessing applications, but what types of industries are really interested and fascinated in having the best, the strongest, the cleanest, most modern database? And then in that, who’s the head honcho of those databases?

Thomas LaRock: I’ll tell you, anybody getting into the analytics space for one reason or another. Let’s talk financials, let’s talk hedge funds, but let’s talk… What do they call the gig economy? Let’s talk startups. Anything where a company recognizes that they can monetize data in general, so that’s a Facebook, that’s a Twitter, where you are the product. A company understands that they can monetize the data they have about you. Those are the people that are going to be interested, especially in the analytics part, and the analytics part is going to come some type of storage and some type of necessity to store data or use data that’s in databases or data lake, data warehouse, so on and so forth. Any company or industry that has got to the point where they employ a team of analysts, business analysts, business intelligence, data analysts, market analysts, analytics, anything of that nature, I think, would be front and center for all of this. But not just them, it’s also research, a lot of medical field. It just goes down to the list. There’s just so much.

Sean Sebring: That’s actually a really good point, because database… In thinking about data in general, it is not just, which I unfortunately silo my thinking into just someone using an application that leans on a database. It’s now a commodity. I don’t like that I’m a commodity because someone tracks everything I click on on my phone.

Thomas LaRock: Then stop clicking.

Ashley Adams: All right. We’ve covered a great many wonderful things about database. Tom, what are you focused on? What are you working on today?

Thomas LaRock: Well, in addition to being head geek here at SolarWinds, which is all the technical product marketing materials that I’ve been providing for many years, I’ve kind of come full circle back to my first love: data science. I realized I had been spending a lot of time as a database administrator, somebody who really helps, let’s say, care for the data in a very tender and loving way. I would care for the data and I would help people with data and the storage of it and stuff. And I realized I wanted to know more about the application of the data. What are we doing with this data? I’ve touched upon different things through my life for stats and things of that nature, but I started realizing that there was this whole data science field that I hadn’t really dived into. So, I started earning some certifications here and there along the way, taking a few classes online.
For the first year or so of the pandemic, I think I took every course that DataCamp had to offer in Python and machine learning and everything. I’ve really gotten back into not just data science, but the data analytics, understanding how to take the data, build a model, make some predictions. For example, the NCAA tournament started, and on Kaggle, they do a competition every year. Can you predict the results of the NCAA tournament, building a model? And the answer is surprisingly, yeah, kind of. You get an idea of who should win. Of course there are humans here, things happen, but I have an idea that, as it stands according to my model, Houston should win actually rather convincingly this tournament. I’ve got the data that backed me up. Will it happen? I don’t know. Should it happen? Could it happen? Yeah, probably.
As a result of all this, I’m enrolled in the online Master’s of Science program at Georgia Tech. I’m going to earn a second master’s degree, this one in data analytics. I haven’t figured out my exact discipline yet, but I think it’s going to be in modeling because I’ve already done the data engineering stuff. I can build you a pipeline easy. The other one is a business focus and I go, “Well, that’s just applying who you’re building the models for.” I’m going to dive into the models and the math behind that and that hopefully I can complete that program in the next few years.

Ashley Adams: It’s fantastic. Eternal student, always learning.

Thomas LaRock: I do just. I’ve got all these books around me that have… I did Six Sigma at one of my previous jobs, which is just stats and it was Minitab. And then I’ve got this thing, understanding the variation and chaos. You’re right, I’m surrounded by data. I have all these little books. I love it.

Sean Sebring: One of those was a really cool pocket bible of data. This literally could fit in your back pocket.

Thomas LaRock: Yeah, it’s a Six Sigma memory jogger. This was the green belt. There’s one for the black belt.

Sean Sebring: Yeah, absolutely.

Ashley Adams: I would make a shameless plug to all the women and girls listening to this podcast to be interested. If this peaks your interest at all, stay in school, stay in STEM, and you could have a career in database or all things IT as well.

Thomas LaRock: Math is so cool. It’s the best.

Sean Sebring: Numbers are infallible.

Ashley Adams: All right. This brings us to the rapid fire question portion of the podcast, where we get to know our guest host a little bit better. Which talent would you most like to have? If you could have any talent in the world, what would it be?

Thomas LaRock: That’s tough. You know what? I would love to play professionally European football. I would love that.

Ashley Adams: Any particular team?

Thomas LaRock: Oh, Arsenal, of course. There’s no other. What other club would you want to play for? That’s crazy. That’s a crazy question.

Sean Sebring: I love that. Okay, next one. Would you rather travel to the past or future?

Thomas LaRock: Do I retain knowledge as I travel?

Sean Sebring: Yeah, you’re just you, same clothes. We’re not doing any Michael J. Fox and start to blur kind of stuff. Yeah, you’re you.

Thomas LaRock: I’m going straight to the past. I’m going to make some money.

Sean Sebring: Nice.

Ashley Adams: There you go. Okay. What is your favorite tech invention?

Thomas LaRock: Well, that’s kind of broad. Are we talking machines or code?

Sean Sebring: Neither. I like to code. Yeah. Let’s say your code. Why not?

Thomas LaRock: Oh, my favorite tech convention in terms of code?

Sean Sebring: Of course, a data person will be able to spin this one to a different direction.

Thomas LaRock: No, it’s going to be PivotTables.

Ashley Adams: I remember when I was first shown a PivotTable and taught how to use them, and it was a finance person that I worked with who utilized them to every degree. I mean, she’s an Excel expert, and I can still remember my mind being blown on many different occasions.

Sean Sebring: Although I think I’m good at math, I had trouble using them at first for very specific use cases for judging people, managing a service desk. We got to see who’s performing the best, weighted categories. Just for judging people, but it was very useful and it actually did a lot. Once you figure it out, it does so much work for you.

Thomas LaRock: Yes.

Sean Sebring: Okay. I’m going to keep on the, I guess, weird sci-fi questions here. I’m going to say if it was an option, and it could be temporary, but would you live in a city in space?

Thomas LaRock: Yes.

Sean Sebring: Nice. Why’d you hesitate?

Thomas LaRock: Well, the question started coming… Well, how far away am I? Are you sending me to Alpha Centauri to live? Even if that was the case, I’d probably be like, “Hell yeah. See you later. I’m out this freaking rock.” God. The distance at first I thought was maybe a blocker, and then I realized it really wasn’t. Sure, let’s go.

Ashley Adams: Sean, [inaudible 00:40:48], when he asked me, it was about a length of time. Let’s say it’s just a vacation, you only have to go for six weeks, I said yes. My answer’s generally yes.

Thomas LaRock: I’ll try anything once.

Ashley Adams: There you go.

Sean Sebring: Turn me into a popsicle and send me to Neptune. I’m just kidding. I don’t want to go there.

Ashley Adams: What time of day would you say you’re most productive? Are you a morning person, a night person, somewhere in between?

Thomas LaRock: I’ll just say in between . I’m more productive at different things at different times of the day. Sometimes it depends on the day, but in between.

Ashley Adams: All right. I’m going to wrap it up with one last question. What is your most treasured possession?

Thomas LaRock: I mean-

Sean Sebring: Don’t say your data.

Thomas LaRock: … possession is really kind of arbitrary. Do we really possess things? It just sort of… Yeah. I mean-

Ashley Adams: Getting metaphysical

Thomas LaRock: Yeah, I mean, a possession, a prize possession? I mean, the easy answer is to say my family, but that’s not really my possession, so…

Ashley Adams: That’s a very endearing answer.

Thomas LaRock: Yeah. I mean, possessions are just things. Nothing comes to mind. I don’t have any one prize possession because it’s just a thing.

Ashley Adams: Fair.

Sean Sebring: Well, you’re definitely getting brownie points because your family was the first and only thing you really said. Hats off to you, Tom.

Thomas LaRock: All right, I’ll take it.

Sean Sebring: Thank you for joining us on SolarWinds Tech Pod. We were joined today by our guest, Tom LaRock. Thanks for joining us.

Thomas LaRock: Thanks for having me. This is great. A lot of fun. Happy to do it again sometime.

Sean Sebring: I’m your host, Sean Sebring, joined by co-host Ashley Adams. If you haven’t yet, make sure to subscribe and follow for more TechPod content. Thanks for tuning in.