Ep 50: Dave Perrill, CEO, Compute North - Colocation for Cryptocurrency Mining

James Thomason: 0:04

It's The Next Wave Podcast, episode 50. I'm James Thomason here with co host Dean Nelson and Brad Kirby. This week. Our guest is David Perrill, He's CEO and co founder of Compute North, and a 25 year veteran of it and InfoSec Dave has been keenly immersed in the cryptocurrency and mining industry and blockchain technology since its formative days. He found and then subsequently sold to technology companies including secure connect an ISP which morphed into a managed security provider and was ultimately acquired by Trustwave holdings in 2012. He also served as CEO of one corporation, a leading point of sale and digital menu board provider to the quick service restaurant industry. Dave has extensive experience in networking, data center engineering, scaling large IT systems and security. He holds a bachelor's in Management Information Systems and an MBA in finance from the University of Minnesota. And he's been a member of the YPO. That's the Young Presidents Organization since 2015. As part of the Twin Cities chapter in Minneapolis. Dave, welcome to the show.

Dave Perrill: 0:56

James. Thanks for having me. Glad to be here.

Dean Nelson: 0:59

I'm so glad you're here. Dave. We've been trying to get you on for a while. You're going to give us some great insights

Dave Perrill: 1:05

in there Dean like a pandemic, you know, those things happen.

Dean Nelson: 1:09

Likely excuse Come on. So can you give us a little just intro to how to you and I met? Yeah,

Dave Perrill: 1:16

you were in Minneapolis or citizen Queen concert? And I think you you know, someone that's actually a big part of that band. Yeah, I heard. Yeah, our friends over at stream datacenters, Robert Fulford and group we're having us over, we've done some work and some development efforts with them in the past, they've also done some energy contracts for us. And so good friends, we got the chance to kind of sit down. And I don't think we really knew how much our backgrounds really, really crossed one another until we dove in a little bit more detail, several weeks later, but it's been fun to see it come full circle and excited for the road ahead.

Dean Nelson: 1:50

Yeah. And that was very fun, because he and his wife came over and we're in a suite watching the citizen Queen open up for Pentatonix. And it was one of the first shows, and I was born in Minnesota. So in Farmington, so we had all these different connections that kind of popped up there. But I remember after the show, we're like, man, we should talk some more. And then he gave me a call. And I was going to come out and see his Bitcoin mines, right, his crypto mining things, and then the pandemic hit. And so then we didn't do anything for, you know, a year. And then we just started jumping back into together and I went and saw his site in Nebraska. That was one of my first trips, after, you know, at least got to the point where I felt I

Dave Perrill: 2:25

could play your first trip if I recall.

Dean Nelson: 2:28

That's right. That's right. See, I came to you day, if I had a priority, it was really important. Really, really important. So tell us about your business. Tell us about what you're doing at compute north?

Dave Perrill: 2:38

Sure. So yeah, I'm a CEO and co founder of compute North started in 2017. The original kind of you know, thesis was to do mining. You know, I discovered Bitcoin mining, I think I was at the money 2020 conference, and Steve Wozniak spoke about Bitcoin and some extensive depth, it really got me interested in it, I was compelled by FinTech and a certain kind of pull back the onion, I got really interested in what was going on on the mining side. And what I found out is it was just compute done in a completely different way compared to where we look at it, where it was all about redundancy, reliability uptime, it was focused on density and cost, and being strategic and thoughtful about the way that you procure and find your energy was first and foremost, to being a successful miner. And, frankly, when I was building data centers, when I was you know, doing the construction, and you know, doing our deployments, energy was kind of a given, right, everything else was the factors, we were looking at fiber connectivity, people, what kind of market were ran, etc. I knew nothing about energy. So I went through my network, and you'd mentioned Young Presidents Organization, which has been just a phenomenal network for me, I met my now co founder, PJ Lee, and PJ comes from 27 years investment banking, project finance all around the energy sector, the last 15 years a renewable and within four weeks, we hit it off my wife, you know, really questioned who I was texting till two in the morning, every night for that period of time, but we both put, you know, some significant sums of investment capital, and started the business. And we found real quickly that, you know, traditional data centers were too costly. And candidly, the ones that were focused on crypto simply just didn't have the professionalism and the background or the capital or the speed to do it, at what we looked for. So we did it ourselves. And you know, Fast forward four years, it's amazing how far the industry has come, you know, the Wall Street, you know, just kind of the household names that have come into the sector and really just given a ton of credibility. And we've just, you know, continually built out with focus on speed, low costs, and above all of this being a trustworthy, incredible partner. So it's been a fun ride and looking forward to the next couple years because they tell people it's 1994 So there's a lot ahead yet

James Thomason: 4:51

and it really feels like 1994 actually feels like 1999 in the stock market. You just sort of log in in the morning and buy some things they go up to three times in price you sell At the end of the day, it's a really great time in the market for stocks as well. But tell us a little bit about your scale. How many facilities are you running? How much computing? Is there? How much power you're drawing? Kind of give us a picture of how big it is? Yeah,

Dave Perrill: 5:11

everything in mining and crypto is really it's measured on a megawatt scale. So in essence, what is the power job? What is the power usage, unlike traditional compute, where you're running maybe 40 50% of the time overall, in terms of your load factor, we're running 9596 97%. So when we're running, we peg it. Our largest facility today is in Kearney, Nebraska, and that is approaching 100 megawatts, I think we're at like 94.5. So we're right around that theoretical maximum of what we can go about. We do have expansion plans going into 2022, where we will put over a gigawatt in the ground with our partners and among several sites. So we do have seven sites under development right now. That's active with our with our customer base with our development partners. So it's going to be a busy year. But again, the industry continues to go up and the right and we certainly you know, they need trusted development and operators like we do I compete.

Dean Nelson: 6:06

So you're only going up by a factor of 10. Dave, I had higher expectations. 100 gig.

Dave Perrill: 6:12

I knew you mentioned that. So it's gonna be exciting here.

Dean Nelson: 6:15

Let me help out with a little perspective. James, I went to the site. And if you think about a data center that's built today, they were saying they're too expensive, which is absolutely right. There's a lot of redundancy and everything else built into it. And all of the histories of why we do that is the question, but just think of it an average of 5 million bucks a megawatt right now for us to build a brand new site. Okay, people have it from four to 10 million a megawatt kind of thing. But in average, probably about 5 million is where things are right now. The modules that are rolling out per megawatt are $250,000 or less per megawatt. Okay, so 1/20 of the cost. Why no generators, no ups. Basically, it's a modular unit that's using outside air, and crankin. Like he said, 97% utilization there. So the reason I'm saying this is that my eyes opened up when I walked out to your Kearney, Nebraska facility, because I went in on like, it's a shipping container, it's modified, it has inlet on one side and exhaust in the other inlet temperature, give me 95, because it was 95 or 96. That day, and 160 Fahrenheit out, it's just cranking and the machines don't melt. Lo and behold, the hard work and hey, take it, it's designed to handle heat.

James Thomason: 7:26

So it is, you know, when I remember a study that Intel did a few years back maybe over 10 years ago now where they, I believe, pitched a tent and actual tent in the Nevada desert and shove servers in there. And they just kept the air moving. Right? So there was no, there's no air conditioning, it was just a lot of airflow. And even under those conditions where ambient temperature and get 110 plus degrees outside the servers were fine. The infrastructure was fine. So it really does appear to be about airflow. And I'm sure Dave, you could tell us more about that. What's the secret? Yeah, I

Dave Perrill: 7:57

mean, the big things are exactly that, right? Are you engineering your CFM to move out the amount of heat that's generated, you can imagine our containers, you know, 40 foot shipping container, as modified for the electrical distribution puts out between 1.8 to 2.2 megawatts, the output of that is all heat. And so the biggest thing that we need to make sure is there's no serialization and no circulation of that heat, that it gets hotter and hotter and hotter. And so having the right egress and the right Ingress, and making sure that those two, never connect has been really crucial the engineering and crucial the design, you know, deems, also, you know, our site and kind of the ways that we've gone down this learning curve. And there's been things that we've tried that haven't worked, there's been things that we have that improve things and like anything, you continually find ways to modify it. We're on our third generation now, and are very confident into design, even towards some of the types of heat, you know, 100, you know, approaching 40 degrees Celsius. And we do plan to have expansion also in Texas, which obviously has even hotter climates than Braska. Longer term, we do see a trajectory to you know, what's happened, I think, in this computing world to drive down the cost, the same type of approach with some other, you know, technologies that I think have been a little more centric, particularly things like dielectric and emerging based cooling. We think those technologies are ripe for disruption. We think they're really primed right now towards HPC in supercomputing type of markets. And we think the types of scale and cost approach of the crypto markets can help drive those costs and make them much more sensible. So those are some of the r&d efforts that we're involved right now. We have some great partners that we're working with on it. And we're excited to start to unveil that here for Sitecore next year.

Dean Nelson: 9:38

And I got a question around that too. I just want people to get an idea. How long does it take for you to deploy a module? How big is it? And when is it operational?

Dave Perrill: 9:48

It's about four days for us to deploy two megawatts. Okay, hold on, pull

Dean Nelson: 9:52

it. Everybody get that? Two point a two megawatt module in four days. That's like land plug on generating right? Crypto hash rates,

James Thomason: 10:04

and that's including that's full of mining rigs, right? That's includes loading the mining rigs. Yep. Wow.

Dean Nelson: 10:09

Yep. When I visited, wasn't it you had 30 megawatts?

Dave Perrill: 10:12

I, you were just starting that expansion phase, the second 70 megawatts in Nebraska. That sounds right.

Dean Nelson: 10:17

Yep. Yeah. And now you're almost at 100 megawatts for that site and a gigawatt in the pipeline.

Dave Perrill: 10:23

Not at one site, just to be clear.

Dean Nelson: 10:26

But in the pipeline, we do have a view

Dave Perrill: 10:27

that we want to be diversified, both to power markets, to generators to even types of, you know, fuel sources, you know, we're seeing obviously, natural gas bike, you know, our whole, you know, thesis is that the compute should happen in the most environmentally conscious, low cost environment at that point in time. And I think over time, we will also move and shift, I don't think all of our sites will be permanent 100 of the time, you know, congestion points get resolved, you know, generators, you know, move, they don't need certain types of assets at times. And so we see a lot of flexibility, continuing being, you know, cornerstone to our business model.

Brad Kirby: 11:04

So effectively, you're hosting the miners themselves versus doing the actual mining, or do also line. Cryptocurrency, the primary

Dave Perrill: 11:11

business of us is hosting and infrastructure. So we do have some development, or pardon me some self mining, but it's more so we keep skin in the game. We know how to do it, we know how to optimize it, enhance it, or software, etc. But it's not our primary driver,

James Thomason: 11:26

selling picks and shovels to miners. Where have I heard that before?

Brad Kirby: 11:29

Yeah, exactly.

Dave Perrill: 11:30

Levi's red. So Wells Fargo, those kind of things. So those are the doubts. Yeah,

Brad Kirby: 11:34

all the all the original guys are moving up. I think you have the right business model. I've actually come across your co founders name before he heard a TerraForm power was actually worked on. It was when I was at Brookfield renewable, has it Brookfield asset management, the parent but they acquired TerraForm power, which I think ever stream which he co founded.

Dave Perrill: 11:55

Yep. Dominic as well. Yeah.

Brad Kirby: 11:57

So So small world in that sense. So I have some,

Dave Perrill: 12:01

again, with infinity global and so just energy construct and celebrate team.

Brad Kirby: 12:05

Yeah, I've come across his name before and literature and in reading. So it's a well,

Dave Perrill: 12:10

we should have him on here too. So

Dean Nelson: 12:13

yes, these two

Brad Kirby: 12:15

ideas away and you you're also on the board of public Canadian blockchain mining company. That's

Dave Perrill: 12:19

correct. Yep. I'm on the board of directors of Hive blockchain. Yeah,

Brad Kirby: 12:22

I know. I've met Frank before as well. So yeah, I'm in Toronto. Briggs. Awesome. So is he in Texas?

Dave Perrill: 12:29

Yeah, he lives in San Antonio. Yeah, but he has 10,000 ideas per 60 seconds. Light. Yeah.

Brad Kirby: 12:35

Yeah. Pretty intense. It's pretty intense.

Dave Perrill: 12:39

He is always go. Yeah, his background. I mean, he's got just a great fundamental background for this from, you know, gold and precious metal mining. That's his construct. And so he takes his investment thesis very similar to that. He's got, you know, a really unique story. I think they've been focused a lot more on the Ethereum side. Now, they do have some bitcoin exposure as well, but Aetherium has kind of been the main driver. They've been really out in front of the ESG standpoint. Adi, Megan, environmentally friendly. They've been in the Nordics. They've just gotten into Canada. No us exposure yet, but you know, yeah, their equation long term.

Brad Kirby: 13:13

That's really neat, because I've, uh, I've known them for five, six years now. So it's a small world, very small world. Yes. To

Dean Nelson: 13:21

me, like there's a new factor here, which is six degrees of separation for next wave. We have we can we can

James Thomason: 13:32

an LA business meeting, like when you go to LA and everyone goes around the room at least twice, saying, everyone they know, we can't we can't do that. We can't know.

Brad Kirby: 13:40

I don't know. I don't know these people I mentioned in passing.

Dave Perrill: 13:43

Oh, you should do a PG on for a future podcast along with Dominic, you'll learn more about energy. I mean, talk about just a huge disruption to what's going on in that world. Right? Yeah, we play within that compute north, but just the huge, massive movement from large scale fossil fuel. But the Hub and Spoke network, to now distributed, intermittent generation, completely spread out. And you think about kind of the compute needs of that. That's where we look at it where we can solve real problems. But there's so many opportunities underlying that and just the energy sector at large, just the momentum. And the movement behind that, I think is absolutely fascinating. So and PJ can speak to it 100 times better than I can, but it's certainly a fun topic.

James Thomason: 14:25

So we got some industry jargon there. What is distributed intermittent generation.

Dave Perrill: 14:28

So think about wind and solar, it doesn't run 24 by seven, it's on it's off, it runs 50% 20% 30%. You think about like a typical coal plant. You run it, you're at 100% As long as you need to be because you have the fuel and you keep running. And so it's a very different way to look at the way the grid works, the way that you consume power, the way you consume energy. You know, Texas is a great case study of what happened in February when a lot of things go wrong in the energy construct and what goes wrong on the grid. Now we have operations He's in Texas. And so I think one thing that's really unique about, you know, what we categorize, as you know, this tier zero compute, is in a situation when the grid needs us, we can and should shut down. And that was an example, we were down for three and a half days. Now, Bitcoin mining is not essential, it's not mission critical. No, you know, customers get it, they understand it, you know, part of what they pay for the reduced rate is understanding that there's that optionality that we can be called for these demand response programs and shut down. But this type of load this type, I think of interaction in grid stabilization, has become even more important as we shift more to renewables, which are, by their definition, intermittent and more distributed. The other thing that enters in the equation is the issue of transmission. Because a lot of times this generation figure with renewables are in spots where the populace or the population and the energy requirements aren't. And if that transmission is not there, then you got huge congestion issues. But what do you need to do to ultimately soak up that energy your use it, so simply not put the ways but one stat that always jumps out at me, which I just think is pretty mind boggling as large as the Bitcoin network is, and you hear a lot of comparisons to the energy use. And these kinds of components. 15 times more energy is wasted every year, because of congestion and inability to put it to the transmission. Wow. So people talking about, like the volatility of Bitcoin, what's more volatile than that energy really hard to store, it's something you need to use right then and there. And that's why the prices can be all over the map. And it's really hard to pin down.

Dean Nelson: 16:24

So in other words, we need a place where we can go use that abundance of energy and not wasted, not necessarily only about conservation.

Dave Perrill: 16:32

Yeah, I mean, I think it's both I mean, I think there's nobody, I think in the Bitcoin narrative that doesn't think that ESG isn't important, right? Nobody's here saying, Hey, we should burn down the world, we should burn more coal, they quit all things equal, or there's a few but what what they might be dealing with their therapists on a few topics, I think, you know, where the focus needs to be, from my standpoint, really is out of the generation of how do we generate the energy? And then more importantly, like, how do we transmit it? How do we use it? How do we leverage it, I think, you know, our type of compute load can play a really compelling factor in that transition, because it's not green or not green, there's a million different shades, there's a transition period, that's going to take time. And I think that you know, people that have just think tomorrow that there's going to be a switch, where everything's gonna be 100%, renewables simply don't understand the engineering, the economics and the markets of some of the things that are at work here.

James Thomason: 17:28

Oil is kind of awesome, right has a lot of energy inside, and it's very stable, you can move it around, it's not easy to replace, I was gonna ask you, you know, there's been a lot of buzz talk, conversation, Buzz news, that kind of buzz about reclaiming the wastes energy from these facilities and generating power from that or using it for other purposes. And I believe, actually, even this week, there was a stock popping in the blockchain space, who I won't mention, because they had announced something like that, that they were going to take the waste heat from their facilities, and then use it to generate more power, is that something you guys are looking at doing it, we've looked

Dave Perrill: 17:59

at that the issue that we've seen time and time, again, is that it's not hot enough to really have much of a secondary purpose here, like a lot of combined cycle generation plants where you know, they have a two cycle, then the end stage and then get, you know, 30 to 40%, more energy on that second stage. Like, you know, Dean pointed out, we're like a 60 degree delta t from our intake to our outtake, so it's not so massive, where we think it would make sense is some sort of hydroponics or something where you need year round heat, but most of the environments that we're in today, have enough heat already to grow the type of agriculture that makes sense. So we haven't seen that fit yet. But certainly long term, I think that's an interesting angle.

James Thomason: 18:40

It's interesting to hear you say that that is bogus, because I like deflating the gas out of some of this hype, right around these types of things. So that's really interesting. Maybe everyone in Texas gets a hot tub?

Dave Perrill: 18:50

Yeah. I think the ESG angle is something you got to pull back the union, right? I don't think all things are created equal. There's I mean, I think when you really start to pull back how this works of what's truly carbon free, what's the carbon offset? What's the renewable energy credit? What is tax equity? What does it mean, when you're generally, you know, behind a nuclear plant versus a fossil fuel versus a wind versus a solar? What about when there's a mix, and a hybrid within that? And again, that's where I think there's a lot of education, and candidly, there's still a lot more definition that's needed of what truly is green? And how do you get there. And, you know, right now, there's a lot of market play. And a lot of, it's a really good marketing of things that, you know, necessarily are good for the environment or not, I don't know if I'd fully agree, but they can be somewhat coated or branded green. And I think that's gonna play out here over the next couple of years, as I think the industry wakes up gets a little bit more understanding and really, you know, the players that are really trying to drive you know, what, ultimately carbon free and renewable and good for the rest of the grid and the rest of the system I think is kind of the the golden triangle that we really want. You know, Yeah,

Dean Nelson: 20:00

I have an opinion about all this, you know, my look at this is that I really love your waist analogy. And if you think about it, I was out of the place called talent, energy, you familiar with them. And they're in Philadelphia somewhere, I think a camera assist Saskatchewan or something like that. And they've got two nuclear power plants. And the two nuclear power plants right are generating 97 98% of the time. And it's 2.6 gigawatts of energy. But they're putting the back out to the grid, but they're at the mercy of the grid prices. So if natural gas happens to go down, they lose margin, and there's no incentives or anything else, they just lose, because of the energy that's going in there. But they're such a big contributor, they have to be on the grid. So they're building a data center campus that allows them to be able to now land behind the meter consumption. And part of that one is their landing, you know, Bitcoin and crypto mining site type things. But that model, to me is where lots and lots of generation of power, you could now go back and balance the grid, you could say, I actually need to put less on the grid, the congestion, I think you're talking about, right? So I'm going to lower goes out, but I want to build I'm still generating and I want the plant to actually do less. So would increase mining within that standpoint. Because you've got the ability to do that behind the meter, instead of getting tariffed all over the place is that's another huge debacle, I think when it comes down to how energy is generated and distributed across the country, in the world. So the point being that they could be almost a heat sink, they can now take all this and then they can drop it, like what you're doing. And I just thought that if you just think of that model, nothing I'm endorsing, you know what they're doing. The point is, here's energy generation, compute, that's not just data center, but it's also taking a combination in this hybrid approach of crypto, it, which is the perfect thing to say I can turn up and I can turn back down, if they need the energy, turn it off.

Dave Perrill: 21:51

Yeah, the marginal cost of you know, nuclear productivity. Zero, right. I mean, literally zero in terms of what they're gonna, how they're gonna produce more energy at that point in time. But I do think nuclear, for whatever reason, is just completely overlooked in the overall portfolio. Because it's not, quote unquote, green enough. And I just think there's still a huge need in use for these baseload types of cases, nuclear plays that end and to your point, Dean, like this type of application can give these a longer life from a lot of these right now are slated to be shut down and decommissioned, because they're not economic. And the prices and the some of the numbers that I was given was that, you know, roughly 10 to 11 cents per kilowatt hour, 110 megawatts, or $110 per megawatt hours were these price at, but 80% of that cost is in regulatory, exactly the actual cost of the actual energy and electricity. So to your point, if you can take away transmission distribution, some of those regulatory costs and bring that in, it makes a lot of sense. And talent is certainly forward looking in that we had several conversations about those sites about three years ago, and it was simply too early. So it's exciting, I think, for the entire industry to see that come around and see a project of that magnitude getting off the ground.

Dean Nelson: 23:03

Yeah. And can you do me a favor? You talked to me before about crypto winters? And you've written through two of them?

Dave Perrill: 23:11

They're called they're even colder in Minnesota than elsewhere. help everyone

Dean Nelson: 23:15

in our audience understand what exactly is a crypto winter? And why do people lose their shirts in that?

Dave Perrill: 23:20

Yeah, I mean, we, I guess a couple of things to point out, you know, we look at Bitcoin specifically really, as a as a platform, right? We see it as a layer eight of the internet, a technology that's becoming standardized, and is seeing innovation really like layer two, right? We see that kind of layer one bars, decentralized, permissionless database, rocking and rolling, being able to do its thing. And we think crypto is one use case. But we see it as a standard. And that's why we get excited about it, that it's something that's going to be here on a long term basis. As far as the like the price and the production of the currency. I mean, it is the definition of a commodity. There's no question about it. Like if you look at it, how its produced, where it's produced all those kinds of components at the end of the day, a bitcoin is fungible. One Bitcoin, by definition should be the exact same as another Bitcoin, which is part of its value, no matter where in the world is produced, no matter where it's mined. For that reason, we have always had a focus on being the low cost producer, by being as efficient as possible about a bill that our construction costs talked about earlier, and also about the energy costs and being able to provide that to our customer base. And the crypto winners is exactly the downside scenarios that you got to worry about. The last one started right after kind of the euphoria of early 2018, late 2017. When Bitcoin first peak right around $20,000. We wrote it all the way down to$3,000. I think part of the ingenuity and just the the pure genius, in you know, Bitcoin, is the fact that it's not just a technological breakthrough, but just some of the ideas behind human behavior and market behavior that was built right into the code and just the way that for example, the way that different faculty, and the ability for you to achieve Bitcoin changes based on the number of actors and the number of people working within the system. And what we saw happen in that crypto winter was exactly what I think was intended to do. The high cost operators are above their marginal cost of production. So they were producing Bitcoin below what it would cost them to make, they turned off. And guess what happened, the pie stayed the same. Everybody got a little more pie. And that continued and continued and continued until it reached a point of equilibrium, where you're still making money, we were fortunate to not just survive, but you know, thrive through that. Do some m&a, you know, do some acquisitions, and like any entrepreneur that's been through these ups and downs, and some challenging moments, when you kind of look into the darkness, and you think it's a rare light, but you're not exactly sure. But you know, we were fortunate to live through that, but I think that the market will go up, it will go down, you know, James, use the point that, you know, 9097 there's absolutely euphoria right now, I mean, if you take a look at what the markets are doing, just generally, we're gonna have corrections, we're gonna have corrections in this market. I think there's some broad based things. And I think you always just gotta be preparing for the worst, hoping for the best. And that's been kind of the game plan to date here with computer Earth, and so far serve as well.

James Thomason: 26:11

Well, the only one better at printing money than Bitcoin is the Fed. So yeah, both in all markets. It's the it's the everything bubble or neighbors.

Dave Perrill: 26:19

TV gets up, James were great.

James Thomason: 26:22

Yeah, I mean, since they're making, they're just printing it, they can give some away. Right,

Brad Kirby: 26:26

and a few other governments out there, they could probably name as well.

James Thomason: 26:29

There's a few, a couple of questions about well, so number one, as an entrepreneur, you've got to be concerned about the next crypto winter. And so I would imagine that you're seeking to diversify your revenue streams have been and that comes to mind that like, these facilities are perfect for other use cases. But the kit inside is not necessarily so for machine learning training, for example, general purpose, GPUs and processors are great for that. But these ASICs in the mining rigs, they're designed to do one thing and one thing only as efficiently as possible. So how hard is it for you to retool one of these containers with new kit and repurpose it to do something else say the next winter comes? How long would it take you to convert these into AI rigs for training?

Dave Perrill: 27:08

Yeah, I guess two answers that, first of all, we're less exposed than most, because we're a host an infrastructure, you know, one of the things that gives us a lot of credit, you know, I guess, confidence going forward is that our customers are now playing five to seven years, you know, their balance sheets are so much stronger than they were even 1218 months ago, you know, they're putting, you know, hundreds of millions, you know, our largest customer will put over a billion dollars of infrastructure in and they're more concerned about being a part of the ecosystem and bringing the token on balance sheet, less concerned about selling it off at that one point in time, like when they were selling the coin to fund their operations. Absolutely, it was a much bigger concern about what was the price at that point in time. But the fact of the matter, they really have that type of capital source to go through these winters and these highs and these lows, and they want to play long ball. And I think that's really exciting for the overall ecosystem, and just the overall volatility of the market. And what we've seen, if your question of, you know, other use cases, we absolutely see the world this way, in that what we don't have is like the actual compute, you're absolutely right. The ASICs do one thing, well, they mined Bitcoin, they mined Mineiro, they mined a theory on like, you know, whatever proof of work consensus network they're on, that's what they do. Well, what we do have is expansive low cost power. We have network connectivity, we have logistics, we have engineers, and we have the ecosystem, the network operation center to run this at scale. And so the idea is to bring other non mission critical, computationally intensive workloads, like machine learning, like graphic rendering, like you know, climate modeling anything surrounding that type of world that we can simply bring in. And the way that we see this working is we simply take away our pots, you know, our two megawatt pods, and replace that with whatever underlying technology we want. If it's GPUs, it's NP use, if it's a large GPU cluster, like that would be our element. And that's why we put it in, we do have prototypes, we do have test cases and use cases of traditional HPC gear I told you earlier, we're going down the dielectric and version cooling extensively. And we see that working both for ASICs and any kind of traditional, you know, rackmount type of infrastructure gear, and we want to be able to play any kind of compute in the world that that makes sense for that we're never going to do e commerce, we're never going to do a mission critical database, you're not going to stream Netflix from tier zero data centers. But if you got a ton of data, a ton of information, a ton of modeling that you want to do, and you want to do it in the most environmentally conscious cost effective in the world, we're going to be the answer.

James Thomason: 29:34

So every entrepreneur has a different answer to this one, you know, your type of business takes money. I mean, you can't just do it in a garage, right? These are facilities with a lot of power, a lot of infrastructure in place to be valuable to your customer. And so when you started, how did you raise money? How much money did you raise, and how long did it take and how many investors did you have to talk to?

Dave Perrill: 29:54

So I feel fortunate that I had a you know, two prior exits, so PJ and I were able to bootstrap out the business for 24 months of its existence. And so some highs and lows in there. I mean, the funny story we wired, one of our I think it was a million and a half dollars for our first miners, two gentlemen that we knew through another gentleman, who was referred to by a guy named happy. And I remember we said, about 60 days, after we were filming, delivering, like 120 days PJ and I looked at one another, like, you know, this could be the shortest lived business in history. So show up, unhappy, fortunate, you know, that they came through machines came through, and we were off and running, we went to external investors after we had proven things out. And what I mean by that is we had gotten through our first Texas location, we had started opening up South Dakota, we felt that we had a business model, and we had momentum, introductory and allowed us to really keep control, and I think make more of the right long term decisions. Now, to your point, we are hitting just a massive capital expansion in terms of the amount of, you know, megawatts and infrastructure and things that we're putting in the ground, and along with, you know, driving down costs on engineering and cost of energy and prospects. The other big area where we need to focus on is our cost of capital, how do we continue to drive that down to near zero to keep us as you know, competitive, and as as efficient in the market as possible. So that's kind of the next big trajectory that we're on. We're fortunate to have, you know, some great tailwinds in the marketplace and some great investor partners. But we're always trying to find ways that, you know, can we do things like securitization, can we do things like project based finance, where we can put these on their own to SPVs, you know, things that they're doing in the traditional data center realm, that have served them successfully are starting to make their ways into this market as well.

James Thomason: 31:43

So for 24 months, you funded this out of pocket yourselves? There's only here, right? Take it away. Ouch. That's very painful. I know, because I've done that. So I relate to that.

Dave Perrill: 31:54

He only called crying like two or three times. It was It wasn't too bad.

James Thomason: 32:01

And your first outside investor, how much was that for? And how long did

Dave Perrill: 32:04

it take you to get our initial friends and family round for about 10 million?

James Thomason: 32:08

Those are good friends to Emily? Well,

Dave Perrill: 32:10

it was I mean, it was probably, you know, between, I want to say, you know, 40 investors roughly, you know, so checks, you know, ranging between 50 to, you know, a couple $100,000. You know, within that, that was kind of that that first round that got us to that next stage. And that article on this, you know, we closed our first $25 million funding round, which was more, you know, institutional grade type of funding. And now we're back off to, you know, like, significantly larger term sheet we expect to announce closing here the next 40 days.

James Thomason: 32:38

Well, congratulations. And that's a great story, because so you didn't initially take a check from traditional venture capitalist, this was all

Dave Perrill: 32:47

I tell Dean, this, you know, we kind of laugh because, you know, he's born in Minnesota, but he's taller California, guys. So he wrote down there, versus the entrepreneurs in the Midwest, or just, it's a little different. And I wouldn't say it's right or wrong, but no, we did not have that issue check.

James Thomason: 33:02

Yeah, my current co founder and I decided that we weren't going to pursue Silicon Valley dollars this time, we're going to pursue funding outside the valley. And so we did the same thing and kind of had a similar story, just not as many on our cap table, you know, find a couple billionaires and that'll, that'll fill out the round pretty nicely.

Dave Perrill: 33:18

I guess I need more billionaire friends. I got to work on that. So

James Thomason: 33:21

yeah, well, there's there's plenty of billionaires in Minnesota.

Dean Nelson: 33:26

Okay. I don't know any in Minnesota, but hey,

James Thomason: 33:30

you can't get to them, you know, at certain times of the year. I think that's the problem.

Dave Perrill: 33:33

Yeah, definitely in a warmer climate for six months of the year. That's kind of the key Minnesota Mo.

James Thomason: 33:39

Minnesota is like the frozen hellscape of Narnia like literally through the wardrobe door and into ice Queenstown. He's

Dave Perrill: 33:45

never been in Minneapolis, there's literally habit trail. So you know, they haven't shell for that go between the buildings, that you don't need to walk outside in the winter months. Because you'll die. Yeah, you won't be comfortable, that's for sure.

Dean Nelson: 33:58

I took my wife to Minnesota for the first time in the dead of winter, to visit my grandfather who was ailing, and we get off right the plane. And then we get into the rental car, and we get to the place where he's at. And the windchill is like minus 30. It's insanity. I take her to, you know, Arctic hell, and she gets out and she just she's California like she's full on California. Let me take it there. Yeah, it's an experience. But my

Dave Perrill: 34:25

dad always talks about the story of getting off the plane the first time he he's from West Virginia originally went to school University, Washington worked at Boeing. He came here for control data and he got off the plane in the snow crunched, and he goes there's no way in hell I work here. There's no way we can always offer to what they said for control data. They accepted it and he felt morally obligated to come here that year. I saw the rest of the story.

Dean Nelson: 34:49

And there you go, and crypto

Dave Perrill: 34:50

for directory. The good news is, you know, it's been a great working environment. We got some, you know, great people, our management team. The bulk is here on Minneapolis, but our CTO is the Bay Area, you know, our CEO is out of Jacksonville, Florida. And so you know, we do certainly have a remote team as well. And you know, we look to hire the best talent wherever that might be. So just

Brad Kirby: 35:12

a question around, you know, everything that's going on with the ban on cryptocurrency mining in China, just as yours a bit. Obviously, that's some big juicy news there that probably impacted everybody in ways maybe that didn't part of that anticipate. However, I think people don't also don't realize how long China has been going on about banning cryptocurrency for so many years. So maybe if you want to give your two cents on the mining ban there.

Dave Perrill: 35:38

Yeah, I guess, Brad, what was weird is that this time, it was different, right?

Brad Kirby: 35:43

There was yeah, there was there was I had

Dave Perrill: 35:45

the same heard it that yeah, one of those things, all of a sudden, what we saw is the hash rates started going down, right, you can see that in real time, in terms of the computation requirements on the network, all sudden started to really, really

Brad Kirby: 36:01

lower than your pickaxe hysterical value, as well. And the next thing

Dave Perrill: 36:05

our phone started to ring, so we were in complete, you know, crazy Build Mode, going into fourth quarter and into q1. But what that did is it really pushed for the acceleration in 2022. And pulling in some 2023 projects. If you look at what we do, I mean, some of the stuff that we do is, you know, 1218 month tech lead times to get on a large scale transformers, some of the mid level voltage equipment, breakers, switches, I mean, some of those things just don't come instantaneously. And so what it really did, it shifted what the supply chain constraint, which was previously in the actual machines, and it moved into the hosting infrastructure. And that's what happened, you know, literally in the the period of two to three weeks, our sales team or business development teams, you know, continue to be extremely busy. But we've got to set expectations, we're like, we're just not going to take orders on things that we can't deliver to, you know, we've we've really taken a very purposeful approach, compute north to be a team that delivers to our commitments, and a team you can trust and like anything in development and projects, delays can and will happen. But we're not going to make promises. We know this. There's no way to meet it. So right now we're taking deliveries into second quarter. And those are things that we're confident our supply chain on your timelines and your deliverables, you know, PPA site locations, all those are lined up so we can make that work. But anything prior to q1, we just had to say no. And there's been some business that has gone elsewhere. But there's also some people taking some really high prices. And I think on some developers that candidly are proven, and I think there's going to be some tears. So we're preparing for that here. Going into 2022.

Brad Kirby: 37:39

Yeah, I think China was responsible for almost 50% of the hash rate of Declan at about

Dave Perrill: 37:45

65. I think us. I mean, they were well over the majority. So I mean, you think about that, yeah, I think what's really compelling and really interesting look at what other network in the world, you know, can take 65% of it offline overnight, and still operate without a hitch. Yeah,

James Thomason: 38:05

that's amazing. A lot of people don't understand this about Bitcoin, how the network responds to differences in capacity versus hash rate. So can you educate us a little bit on how that works? Exactly? What happens when you add or subtract capacity from the network? Yeah, I think that the best

Dave Perrill: 38:21

way to think about it is, you know, difficulty. I guess, let me take a step back, Bitcoin incentivize miners, to run the network by paying them out in two things. One is what's called a block of rewards. The other part is called the incentivize of the transaction fees. And if you fundamentally step back, like what makes these people spend millions and hundreds of millions and, you know, getting to the billion dollar range on equipment on gear to run this network, it's the fact that they get paid out in Bitcoin, and have an economic incentive to do so. And so those block rewards and those transaction fees are the same for the most part. With the efficiency, there's a the transaction fees shift, every single block, which is set every 10 minutes, and that's you know, relatively flat over time is you kind of see it as a percentage. And the blockchain is every four years. So in terms of what actually gets paid out for the miners, the genius behind bitcoin, is that what's called difficulty or the number of people that's on it only affects you as a fraction. So if you look at the Bitcoin network as an entire hole, and you look at it kind of like a pie, let's say you're if you have 5% of that hash rate, over time, you're gonna get 5% of those rewards, if you're 10% of the hash rate, 10% of those rewards. Now, if someone comes on board and adds 10% to the network, by definition, you just lost 1%. You know what I mean? And so, but it goes both ways. So it works. When the additional computing power comes online, you get less and as additional power comes off, you get more that's a reset that's done every two weeks and just again, part of the genius of The Network again, what I talked about is the fact that this was thought through of how does this scale? How does this grow? How does this not look like when there's two laptops running on this thing, but literally, you know, billions of dollars of equipment focused on maintaining and keeping this network secure. And at scale? You know, again, the one thing that Bitcoin has it no other decentralized network has, is, you know, 1213 years of track record, and that count for a lot as you know, what IT and technology got to make something 10 times better overcome that network effect. And I don't think we're there.

Dean Nelson: 40:31

You know, that brings me to another question, because you talked about 10 to 12 years worth of history. There's a website that I think all of us know about will share with our audience. So it's the CBE CI, it's basically the Cambridge Bitcoin electricity consumption index. So if you go to CB e ci.org, you'll see this over time, like since the birth of Bitcoin, and in that you see what's happened, really, since the growth in 2017. That's where things really started to pick up when it comes down to the overall consumption of energy to be able to now do Bitcoin. And you can see the trends that are happening. In other words, the China decision has a noticeable drop in overall consumption. And just so people get an idea right now there is 112 terawatt hours of consumption. That's 12, almost 13 gigawatts of consumption for Bitcoin alone. That's today. That's like right now they're doing real time hash rate tracking, but that was 141 terawatt hours, just three or four months ago. Right. Wow, China made this decision. And they dropped 3040 terawatt hours out of the actual math that's going on and all these machines. And just to put that in perspective, right now, the total amount of consumption for digital infrastructure, like globally, is 594. terawatt hours. So 1/5 already of everything. Yep. So and again, it went down 30 terawatt hours just within four months. And it's coming back up. And if you look at the trend on the charts, here, it's really interesting to watch that you can see, you can see the crypto winners as well, things change right in it.

Dave Perrill: 42:17

And to things like people that are running at high price, energy prices that were selling, had to turn off and all they're all turning it back on the same thing goes for, you know, previous generation rakes. If you think about like the efficiency game that's being played here. New generation rigs, by definition are going to be more efficient, use less energy to produce more Terra hash, and therefore more Bitcoin and be more productive. But in these kinds of scenarios, like you can go run, you know, s nines, which came out in 2016. And they're still very profitable. And we're seeing a market for that, you know, one of the things that we've also started kind of the genesis of is what we call life cycle mining is where we do see a future where the second and third generation assets, go to wind farms and go to places where maybe they only run 20 3040 50% of the time. But I think those assets are depreciated, they already work. And they're already profitable. But if they could run, let's say, for example, in West Texas, some of these wind farms will produce it literally negative energy rates. And the reason is because of tax equity. So there's some some things that really kind of changed some of these markets and create some some strange incentives. But at that point in time, and if you're paying zero, you're actually doing better for the underlying asset. But maybe they're only running for those six hours, and then they shut off. But again, instead of throwing the asset in the trash, not using it, those types of use case scenario, so we're pioneering some of that, I mean, it's still early innings, you know, we have seen the last couple, you know, sessions every three years, the old stuffs cut out the door. And you know, as people continue to invest and improve, and but as the Dean's point, is really starting to reach institutional, industrial level scale. And that's part of the fun, exciting part of it. You know, I don't know if the decentralization is going to work at a hobbyist level or even running at your house anymore. But the scale and the momentum behind it, I think it's good for the market overall.

James Thomason: 44:05

Do you think bitcoins gonna replace money?

Dave Perrill: 44:10

Ah, there's a great book on this called the Bitcoin standard, highly recommend it. I mean, just really lays out kind of the, you know, the monetary component of this. No, I don't think it's going to entirely replace money. I think it'd be really interesting to see what part of a reserve currency status Bitcoin takes on in the future. And I think that's like the kind of the end all like if it becomes a reserve currency, I think there's a lot of emailings is good for the US. But I think there's a lot of other nations that are really interested in not being pegged to the US dollar. I mean, even just think about the way that like oil futures are dealt with today. There's just a lot of markets that I think would really benefit by being of the US dollar. And I think Bitcoin presents a pretty interesting use case. You know, when I fundamentally step back on like, what does Bitcoin fix it fixes trust. That's what it does. You don't need a third party intermediary, whether that's a bank, whether that's a lawyer, whether that's a county clerk, nobody needs to transact this for you have an intermediary that all sides agree on. And I think some of those transactions like if that starts to change, things are going to get really, really interesting really fast.

James Thomason: 45:20

Every day, there's a new crypto invented or particularly now NF Ts are kind of going nuts. Right? And I've been thinking of introducing some NF T's and my own, I don't know, picture of my foot. I don't know what will sell but something will sell for at least at least 50 bucks worth of Aetherium I think but okay, so And of all these Kryptos everyone thinks that they're gonna replace Bitcoin everyone thinks they're gonna replace Aetherium. Who do you feel really bullish about? This is not an audience? Should I take this as advice to purchase anything? Or trade any kind of anything, but out of all the new emerging Kryptos do you think has the best shot at displacing the lead boys in the pack right now? And I think maybe you have some unique visibility into this because of who is who your customers are?

Dave Perrill: 46:05

Yeah, I mean, when you start to look at the entire ecosystem, just the amount of innovation, the amount of just unique ideas that are out, there are nothing short of just fascinating. And I just love I mean, I think there's 27 Different now types of consensus algorithms, like how do these these centralized databases agree, like, just, you know, some of these unique ways to mining and these approaches, I just think are great. I don't think bitcoins going anywhere. Like I said, I think it's a it's a standard, I think it's fair to say, I think we're gonna see a lot of innovation on layer two. And I think there's going to continue to grow, I don't think there can be 12,000 Different cryptocurrencies. Just like they're going to be a byte count and amazon.com and flus.com, back in the in the late 90s, right, like, you start to take some of these analogies of history. And I think that that is the most relevant thing. So the ones that I'm following that I think are really interesting, I think Charles Hoskinson, and Cardano, I think the ADA tokens, I think, what's going on there, again, whether it's a good buy or not, I again, I don't know how you value some of these things. But the underlying technology of what they're going about with the consensus algorithm, the way they're approaching the technology, I think, are really cool. I am a big fan of Aetherium. But I think one of the things to understand is that it is centrally controlled. I mean, I think, you know, deflation, inflation components are there like they are with Bitcoin. And they a lot of people don't fully understand that instead of, you know, a fully decentralized network, you're literally at the core group of developers and what they decide, and you know, they delay a logic bomb that's gonna affect you. And if they put a new EIP in place, that's gonna affect you. You know, polka dot, I think is really intriguing. And some things they're doing. You know, I think outside the box, if you guys follow helium, for, like, fantastic, great use of, you know, different ways to approach it, right? How do you build a worldwide wireless communications network that anyone can use and mine in a unique way and create incentives, I just like, that's one of the business models that I just think is so fantastic. And just so well thought through, I think they're going to be extremely successful. So, you know, we also in our facilities, you know, we do Mineiro, we do Z cash, you know, we do light coin, you know, some of these other secondary markets that are focused on privacy and some of these certain applications. I think that we'll continue to get innovation over time. But I think we're 92 to 93%. Now, Bitcoin, and I don't see that changing in the months ahead.

James Thomason: 48:26

None of that was investment advice.

Dave Perrill: 48:29

Absolutely. I was at Creighton University last week doing a talk on cryptocurrencies and blockchain. And the question came up for you know, as a young student, you know, where should I invest? And my point blank answer was buying cryptocurrency on Robin Hood is not investing. Right? And I said, if you're gonna go buy a coin, and hope it goes up, that's not the definition investment, what I think is investment, learning about the technology, learning about the ecosystem, learning how you can actually put it into practical use, and then applying those skills. I'm like, That is the best investment that each and every one of you can make work. Otherwise, again, I think, if you're hoping the coin is going to go up, maybe it will, maybe it won't, and I think that's a hard way to invest.

Brad Kirby: 49:11

I get asked that question more often than I get asked what vaccine I took. So I'm sure. Definitely, it's the number one question I get from people. And I almost always but those companies that you mentioned, I actually, they're all pretty good. Concur.

Dave Perrill: 49:28

value prop, right, like there.

Brad Kirby: 49:30

Yeah. And I'm not saying from an investment perspective, just from the actual fundamental project perspective. Yeah.

Dave Perrill: 49:36

Agree. Same view, Brad. Yep. Well, Dave,

Dean Nelson: 49:39

we end the shows with one question. And that's about predictions. You want to give us a picture of what you believe the future is going to look like in five to 10 years.

James Thomason: 49:49

And be specific. Where's Bitcoin next year? Price one year, price wise, what's the prediction?

Dave Perrill: 50:00

Yeah, like I said, I mean, I believe in Bitcoin, I believe in the ecosystem, but we're around it right, we're helping provide it an environmentally friendly, cost effective scale. And that's our job. That's our role, as far as you know, what the price should? And can will, there's so many other factors at work here. You know, we'll see. And again, that's where I think if you're investing to it around the ecosystem that necessarily is buying coin, that may be my advice. I think the future, you know, I'll take it from the standpoint of compute north, and you know, the cryptocurrencies and kind of where I see this coin, I see, you know, crypto becoming more and more ingrained in everyday life. And I think it gets to the point where you don't think about it. And what I mean by that, I think the more obscure the technology is that it's an underlying penning and really a platform play, I think the value is really going to be created at the layer to the things like Lightning Network and some of these other innovative projects that I see coming to bear, I think it's going to be easy to use, I think people aren't going to think about it. And I think just the whole idea of blockchain and crypto is going to be somewhat behind the scenes, I think we're maybe the more visible piece will be the elements of compute getting much lower cost, much more environmentally friendly. But as we know, what's going on and just the world in general, at just incredible scale, like the need for data centers, the need for compute the need for all this is going to continue to grow and scale. And I think the way that we can help proliferate, you know, renewable generation, I think nuclear is a great answer as well. You know, everybody, you know, nobody wants to screw up. You're right. And I think, you know, our industry is no different insofar as we can help accelerate that be good citizens. I think that focus is going to be just front and center. And I think the companies that are focused on that, do it right, and don't just give a talk actually followed up with action and are doing it the right way are the ones that are going to thrive. We hope to be a part of that selection set. So we're excited for the future. And you know, once again, we appreciate you having us on today.

James Thomason: 51:59

Deeper Oh, thanks so much for being on the next week podcast folks if you enjoy shows such as this one, where we bring in the best and brightest minds in business and tech please do give us five stars it helps us grow our audience. This podcast is sponsored by infrastructure Masons whose uniting builders of a digital age. Learn how you can participate by going on the web to imasons.org That's iMasons dot O R G. And by EDJX edge cloud made for planet scale. Create smarter, faster apps, websites and data pipelines on EDJX secure Global Edge platform visit us on the web at edjx.io. That's edjx.io.