Darin Briskman and Josh Bartliff with Imply

SUMMARY KEYWORDS

utilities, data, imply, open source, database, real, company, josh, happening, world, industrial, talk, druid, Darin, running, analytics, apache, platform, technology, cases

00:04

Welcome to the Industrial Talk Podcast with Scott MacKenzie. Scott is a passionate industry professional dedicated to transferring cutting edge industry focused innovations and trends while highlighting the men and women who keep the world moving. So put on your hard hat, grab your work boots, and let's go. Alright,

00:21

once again, welcome to Industrial Talk. Thank you very much for joining and yours continued support of a platform that celebrates industry professionals all around the world, you are bold, you are brave, you dare greatly. You innovate, you collaborate, you solve problems. And that's why you are changing the world. As you can tell by the buzz in the background, we are at DistributeTech here in Orlando, Florida, a great venue, a wonderful opportunity to meet incredible individuals that solve problems that understand the dynamic world of utilities, it is happening. And here is the place that you need to put this on your calendar, just bottom line on a line. All right. Speaking of problem solvers, we have to imply as the company, the two individuals, Darin, and Josh are in the house, we're going to be talking about real time analytics and why that is important. So let's get cracking. Yeah. Especially now, in the world of utilities, absolutely. You better have real time stuff.

01:27

It's all about real time, you

01:28

know, when things start to fail, and they always do, they tend to cascade. And if you don't get in front of that cascade, it's going to crush you. That's why Real Time matters. He doesn't help me to know, you know, here's your report that says 10 minutes ago, I was about to have a big outage.

01:47

No. And and I don't like it as a consumer of power, a proud consumer of power.

01:53

Let's put it that way.

01:55

It's very hard to podcast without power. That is very hard

01:58

to do a lot of things without power. Yes. That's being the sort of at the bottom at the bottom. But it's so true. All right. Before we get into that conversation, I'd like to start with you, Darin, for the listeners out there. Give us a little 411 on who you are. Sure.

02:17

So I'm Darin Briskman. I lead Business Development at imply talk more about imply I'm sure a bit later. But we're the company the database for when real time matters. We create and sell high speed real time analytics databases. But

02:33

what about you? What about your your street cred? I have did you come from

02:36

a weird background? So I started, I started my career at NASA in Texas.

02:42

So God, no wonder you don't say that, because I'll get all distracted by NASA. You know, NASA

02:48

was great. I will say for those who know that your way around Galveston, Houston. Clearly because the most misnamed geographic feature strike people, they were a great guy to work on that back then shuttle and station a whole bunch of interesting things. And then ended up through a process I don't understand. I started there doing physics and ended up as a running it. And ever since then, I've been doing stuff with data and databases. long streak at IBM. founded my own startup sold it spend time at Couchbase Amazon Web Services snowflake. A couple years ago, I came over to apply because I was so excited about real time what this company

03:28

is going to love a Josh Yeah. Background. Yeah.

03:32

25 plus years of customer facing sales. Background, mostly in the Data Integration space, real time data integration, workflow optimization, and spent time with companies like Tibco, and then Salesforce, and now with imply right, so it's all real time, right? That's what matters. That's where I see the future going streaming data, running analytics on real time events. So really kind of the tip, tip of the arrow in front of customers having conversations every day. All

04:03

right. Talk about why real time. Data Analytics is important in the world of utilities. It's pretty simple. It's real time data is important with a lot of things. Absolutely.

04:18

But when we talk, you know, industrial uses, and that whole intersection of OT and it really gets down to everybody needs analytics. You've all been doing analytics or agency. The real time is the real definition is when I need to know something and under a second. So there's a lot of stuff in analytics that's very important to running a utility or transmission or distribution or metering or whatever. You know, if I'm doing a metering and doing somebody's bill, I don't need to know that within a second. Right. That's something and so you know, that's something I can figure it out in an hour I figured out in day five, but real time is things I need to know to actually make my operation run effectively and when decision Cars need to be made often in timescales that humans can't make those decisions. So I need to understand what's happening and do something about it in less than a second, often, much less than a second. And we see that all the time across transmission distribution, other pieces, where you need to adjust loads, adjust flow, change your grid planning, do things, react to a lightning strike, and react to that idiot touch that truck driver just rammed his truck into your substation, all those things that happen. If you can't respond to them in real time, you get a cascading failure. That's what we all want to avoid.

05:38

So, Josh, how do we, where do we even begin? Is there a sort of a prioritization on? Yeah, we want to see everything in real time. That's great. I guess it's possible. But But aren't we already doing something similar? To take us through sort of a use case?

06:00

Yeah, that's a great example. I mean, that's a great question. I mean, I think a lot of the big companies, right, a lot of the utilities have traditionally been technology, laggards, right. I mean, a lot of the back end systems they've had, they've had these systems in place for 2030 plus years. Right. So those of

06:15

you listening, Scott is using a technique called sarcasm. Yes.

06:21

Because nimble is not the word I would use.

06:25

Yeah, so. And, you know, 20 years ago, the world was not real time, right? I mean, cell phones, early 90s, right, you can imagine and the advancements in AI and ML, just in the past five plus years, so the world is definitely trending away from, Hey, it's okay to run a report out of a database, and then come back in 10 minutes, right. And the world operated that way, right. But in today's world, it's all about sub second response times at scale. And that's where the utilities and just a lot of the other pure IoT companies are headed. And so they need to understand the information in the moment, we've seen some clear examples just out of Texas. I say it's my co home state, where they had, you know, pretty bad event, right? And,

07:08

as, but they want to be called ERCOT. They want to be an island they want to be, yeah, their own little power

07:16

Island. Yeah. And, look, there are weather events, right, we're seeing more and more of that happening in the world. And so it's just an absolute that, you know, running systems that are 20 years old, you have to modernize that you have to start at the data layer, you have to be able to expose that data and get that data into something that you can make sense of. And so I would say a good example, it's we have one of the one of my lords. So I said, I'm customer facing, and I have a set of large enterprise global accounts. And one of them happens to be here at the show Honeywell, they've got a Honeywell connected enterprise division that is becoming a modern software platform company, right. They're taking a very industrial based company, and they're modernizing it as a software platform. They've created something called the Forge plus platform. And so they're showcasing their utility usage and their metrics and their platform around utility observability observability, observability. And anomaly detection, say that really fast without a southern accent, that's your that's your responsibility. I have no, but so they're showcasing that and as part of that they embed, they embed our company and our real time analytics database, druid, Apache druid, to showcase to their customers who are running, utilities, and anomaly detection, but they want to be able to make decisions in sub seconds, right? They want to be able to run reports and queries against millions or even billions of events that come up these sensors in real time. And they want to make sense of it. Not five minutes from now, they want to make real time decisions in the moment. And so that can be from, you know, preventative maintenance, or it can be from Hey, we have an anomaly that is way out of scope that we need to take an immediate action on or reroute flow to a different station or you name it. But it's all about real time. It's all about right now. dashboarding dashboards have their place, it's just not necessarily in this season.

09:07

Darin, it's one thing to be able to collect the data just as okay, I can grab the data, grab the data, grab that tsunami of data, it's coming in and but it's still it's how do I make sense of that data? Or we start talking about you mentioned druid, the the product at patchy druid if I if I How does that How do you manage that tsunami of data and make it relevant, make it actionable, make it tactical.

09:37

So it's kind of you're hitting really on what the what the whole problem here is. So and one of the things that makes the, you know, electric world a little, you know, a couple of things I should point out one is, this isn't like some you know, e commerce company in Silicon Valley where if everything goes down, somebody gets their T shirt an hour later, right? If we mess up in this industry people die. Yeah. So we can't be as nimble, perhaps, as we would like to be sometimes, because we, we have to mitigate the risk to make sure that the power is on there. Within that, that also means that we have systems of a wide range of ages. So a lot of mainframe still running out there. Can you believe that? Yes, I can. Because you know what mainframes are not fashionable, but they work. Yeah. And it costs too much, but they work. And eventually they get replaced. But then we also have all of these, you know, smart meters, and we have some SCADA systems that go back to the 80s, and 90s. And we have smart meters or new Stata, that's brand new, we got to get all this stuff to work together. Because the real issue when we talked about, you know, the large amounts of data, you're talking about volume, and volume is a big deal. You also the problem of variety, you have a lot of different types of data, that all have to be pulled together to make sense of things. And then of course, you have velocity, you have data that's really interesting right now, an hour from now who cares? It's time, so I've got to be able to use that correctly. So when we put together you know, this velocity and, and variety, and volume, that's what really makes it the interesting of the big problem. So how do you actually pull that together? Well, one is you need a database, right? That's where the data goes to be seen and pulled together. And you need to have a database that can handle both the streaming data that's coming in from those latest devices. Right. So for those who are not deep into this, a device will generate what's called a data stream, which is essentially just a bunch of events, right? So measuring temperature, measuring status, measuring flow, whatever, and then sending that information once a second or 10 times a second or whatever, to someplace coming in. So I've got all the streaming data, then you also have to be able to mix that with more traditional data, the stuff that comes out of the mainframe once an hour or once a day, and then be able to make things make sense. So we mentioned Apache druid. So this is open source software. It's public, it's free. It's open source. Yeah. So Apache foundation is behind many of the biggest open source projects that are out there. There's two big foundations a Linux Foundation, the Apache Foundation, Apache foundation tend to be more about data. Linux Foundation is more about the operating system. They overlap. And you know, everyone cooperates, the people who created a path Apache druid, then later founded our company imply, so that we could provide commercial support, we could provide database as a service, we can provide the things that you can't get out of the pure open source. So there are 1000s of organizations running open source Apache tried, we're happy for them to do it. There's a subset of that that used to pay us to get the commercial support the security and the as a service

12:52

options, but still, we're at a point where I still get the data, you still can collect the data, the abate? How do I make sense of the data? What's What if I'm looking at real time less than a second? I want I want tactical insights? Are you are we talking about sort of that an algorithm of some sort to define that what it helps us understand

13:14

generally, there's, there's three things that you would be doing with this once I've got the data in that consumable place of the database. So what is dashboards, right, and real time dashboards to deal with sports that are telling you what's happening.

13:27

Good, yellow, watch red, bad. Yeah. And you know, pictures

13:31

of your grader, rather pieces and all that, all that showing of what's going on. But that also allow you to drill in, okay, this is yellow y double click on that right, zoom in there, very rapidly, we're going to take it a little bit technical, these are what are called aggregates, and roll up. So I'm putting together a lot of data. But I still need to get down to the granular, right. Okay, where's this problem? Why is this yellow? Dive into you find? Uh huh. You know, that's what that problem is. Let's, let's fix that. All right. The second thing that comes up here are analytics applications. So this is when people write applications, right develop code. Some of these are packaged services, which many people here at distributech are selling. They are. And some of this is you. This is why you had developers who write who write the code are usually pretty simple applications. Because the tough part really is the data management, but still how you present that and do it. And that leads to the third use, which is another sort of that application, which really cold real time decisioning which are the things where you take the human out of the loop, usually because humans just aren't fast enough. Right. And, you know, to pick a different industrial area, one of our great customers is Halliburton. We're using these technologies for wellhead management. Right. Well, if there's something wrong with the drill, you've got less than a second to figure that out. Yeah. And either turn that drill off or change the lubricant or do something before you have either a fire or an explosion. You can't wait for for a human to look at that and figure that out, humans just aren't faster. So this is where you have real time decisioning, where the humans create the rules that say, if these rules are violated, do something. Now, don't wait. And so these are the ways that people actually use this. So the that first step is getting the data in a place, you can do something with it. But that's usually teamed with your plan to what do you do with the data in

15:22

this world, and we've had a number of conversations specifically around efficiency. And, and to gain or to achieve some of the standard or to achieve some of the requirements that are being placed on utilities, there has to be this commitment to greater efficiency. And that's, to me is a data conversation. It's a speed, it's a, there's a there's a rapidness behind that.

15:52

That one really base to that is being able to monitor and work with your infrastructure and know where failure is going to happen before they because it is 10 to 20 times cheaper to fix it before it breaks. Yeah. Right. Because then you're doing it during the day, it's scheduled, your crews are out there getting the work done, it's not two in the morning, or something just failed during a storm. And you know, that's across everything that's generation as transmission, that's meters, that's the whole end to end world. And again, that's true in a whole bunch of other industries as well, is that predictive maintenance. So this is when the world of machine learning comes in. So not only show what's there, but being able to do algorithms that say this is what's likely to fail the next week, this is what's likely to fail in the next month to get again, get in front of the problems before they happen. And then pure operational is also you know, dealing with electricity, except for that very last. Last leg, there's usually a whole bunch of options for I route things through my grid. And what's the best option right now might be different 60 seconds from now, yeah, probably will be different an hour from now, because of you know, different load levels and different usage patterns, industrial versus residential, and everything else. So being able to monitor those things and make them happen automatically, is another way to pretty radically improve efficiency. And this is where, you know, it can really help ot for being able to do these kind of analytics and make them work in a way that is automated. And again, we see this model, we see it in getting into utilities. Now, you know, to name someplace else that has kind of the same problem is thinking someone like Netflix or Disney plus, yeah, which have many routes to route that copy of frozen plus to your TV screen, right? And they're constantly looking at that boat to make sure it's performant. So that you don't get a bad viewing experience. But also, within performance, what's the cheapest way to get it to you, so that they can minimize their costs and have the resources they need to operate a business. And, you know, when this, this is the kind of world that you know, we you don't normally think of Netflix, you know, next to electrical utility. But while the use cases are different, the actual algorithms and technologies are very similar.

18:07

Wow, that's interesting. All I care about is I don't want to see a little circle thing, just trying to think through it. That's Yes. That's about the extent of my knowledge that exists here. Little circle thing today

18:18

is a highly technical term for that. That's the circle of the spinning wheel of death. There it is. Wait, yeah, nobody wants,

18:23

nobody wants to see that. Josh, let me ask you a question. With all of this going on, where do you see? Where do you see imply going? What's What's the latest? Put your future hat on?

18:35

Okay, question. No. Look, we talked about open source earlier, open source is such a great breeding ground is a gathering of people like minds that really care about technology and growing technology, but want it to be available to the masses. And so, you know, I'm very familiar and have a lot of experience in that open source world. And, and so I believe in it wholeheartedly. I think what implies done and with our four founders of the original Apache geode, open source database, is they've just recognized that as more people have have access to the power of open source technology, that they can start to do some pretty cool things with it. And it grows really fast. And you get into real enterprise revenue bearing use cases pretty quickly, because it's so easy to use. And you can find yourself a little bit out of your skis if you don't have the support of a company like imply, and there's others that that create these really, really efficient managed services around these open source technology. So you know, it's only going to grow open source is going to grow, but scaling and these use cases that are revenue impacting are happening in sub second. We're delivering these use cases much faster now. So it's gonna grow really quickly.

19:45

You want to know a little tidbit of information. I interviewed the guy that came up with the term use case just FYI. Yeah, I'm that important. Nice. And it was funny because he just said, Yeah, I was just trying to figure out what Have a word that just says, Hey, we're going to be to use case that's that was all it was nothing nothing. Nothing beyond that. Thanks for having us. Hey, how did they get a hold of you Josh?

20:13

LinkedIn Josh part left on LinkedIn is the most is the fastest way Josh dot Bartlett that imply dot O is my email address.

20:20

Excellent Darin,

20:21

Darin dot Briskman at him fight at IO just come visit him play.io And you'll find information on this.

20:28

All right listeners, we're gonna have all the contact information for these two gents. Darin, are you out on LinkedIn? Of course. Okay, there it is. We'll have their LinkedIn will have their stat card out there. Most definitely. All right, we're broadcasting from distributech here in Orlando, Florida. It is once again as you can tell by Darin and Josh, problem solvers everywhere your step your your stubbing your toes on them. Anyway, we're going to have another great conversation shortly coming from stay tuned, we will be right back.

20:55

You're listening to the Industrial Talk Podcast Network.

21:05

All right, that's Jared. And Josh implies the company DistribuTech was the event. And if there is ever a need for what we call, high speed, real time, data analytics, it's in the utility space. I love what they're doing a lot of knowledge there. It's just getting better and better all the time. As you can tell by all of the conversations we were having at DistribuTECH and other events, there's a lot of innovation and a lot of solutions that are happening out there. And a lot of individuals that would love to be able to collaborate. Big time, you got to reach out, find their stat card out on Industrial Talk and make it a point. It's on your to do list to make it a point to connect with these two gentlemen. All right. Industrial Talk is here for you. You have a podcast, you want to be able to get it. Get more attention, put it out on Industrial Talk, I'm here for you go out to Industrial Talk.com It is a platform that is dedicated to education, collaboration, as well as innovation. It's all there. Be a part of the ecosystem. Be bold, be brave, dare greatly hang out. We're gonna have another great conversation shortly.