The race to build AI infrastructure is often framed around the availability of power, chips, and capital. But another critical constraint is becoming impossible to ignore: skilled labor. In this episode, Shayle sits down with Raymond Hawkins, chief customer officer at Compass Datacenters, about why a shortage of qualified electricians has emerged as one of the biggest bottlenecks to data center construction, and how the industry is responding.
Hawkins explains that labor shortages aren’t preventing new data centers from being built, but they are driving up costs, extending project timelines, and forcing developers to compete aggressively for experienced electricians. As AI demand accelerates, companies are increasingly moving workers across the country, housing them near construction sites, and relying on a traveling workforce much like the healthcare industry relies on travel nurses. At the same time, demand is surging for highly specialized professionals capable of installing the sophisticated electrical systems that modern AI data centers require.
Hawkins also explores how modular construction and prefabrication are reshaping the way data centers are built, and how, as data centers expand into new markets, developers must become more transparent about their projects to earn a community’s trust.
Shayle and Raymond discuss:
- Why electricians have become one of the biggest constraints on AI data center growth
- How workforce shortages are changing hiring, training, and construction practices
- Compass Datacenters’ partnership with community colleges to develop skilled labor
- The role of modular construction and prefabrication in reducing labor demands
- Strategies for earning community support through transparency, workforce investment, and responsible infrastructure development
Resources
- Catalyst: The State of Play of Data Center Development
- Catalyst: AI Scaling Pathways: On Grid, On Edge, Off Grid, Off Planet
- Latitude Media: As community scrutiny grows, data center developers turn to on-site power
- Latitude Media: The mechanics of data center flexibility
- Latitude Media: The off-grid data center fantasy
Credits: Hosted by Shayle Kann. Produced and edited by John Sheehan. Original music and engineering by Sean Marquand. Stephen Lacey is our executive editor.
This episode of Catalyst is brought to you by ENGIE, the smarter energy supplier. ENGIE doesn’t just provide the power to run your business — they supply the energy to move it forward, with reliable, flexible solutions built for what’s next. Learn more at engieresources.com.
Catalyst is brought to you by EnergyHub. Peak season puts every grid to the test — and the utilities that pass are the ones that built flexible capacity before they needed it. EnergyHub works with more than 170 utilities to coordinate 2.5 million devices and 3.4 gigawatts of dispatchable flexibility through a single platform designed to perform when it counts most. See what that looks like at EnergyHub.com.
Catalyst is brought to you by Bloom Energy. Bloom Energy fuel cells deliver affordable, ultra-reliable onsite power for hospitals, utilities, and data centers – at speed and at scale. Learn more by visiting BloomEnergy.com.
Transcript
Shayle Kann: I’m Shayle Kann. I invest in early-stage companies at Energy Impact Partners. Welcome to “Catalyst.” So I, I’ve noticed that I keep coming back to this question of What is the current rate limiter on AI infrastructure growth?
As I’m sure you’ve heard if you listen to this podcast, I think the main candidates are demand, in theory, nobody really picks this right now, chips, power, and labor.
The last one people mention a lot, but then I’ve found kind of move right along, whereas the others, chips and power, get a deep dive worthy of their rate limiting status. I think that’s because labor feels to be maybe the most intractable of them all. To a first order, the AI industry can kind of throw money at the other problems. They can buy chips from AMD instead of Nvidia. They can try to convince TSMC to expand capacity. Or on the power side, they could buy gas turbines to throw behind the meter from GE or Siemens or an air derivative supplier, or they can sell out Bloom’s fuel cells for years and years and years. Not to say those things are easy, but when it comes to labor, it’s sort of a trickier problem.
To be clear, lots of players in the market are trying to do something, workforce investments of one kind or another. But I think it’s just a little bit outside their comfort zone. Anyway, the result of this is that I think workforce, particularly for electricians, is maybe the least appreciated, but certainly one of the most important bottlenecks in this market.
And I wanted to dig into how it’s actually manifesting and what one might do about it. For this one, I brought on Raymond Hawkins, who’s the CCO at Compass Datacenters. That’s coming up next.
Shayle Kann: Raymond, welcome
Raymond Hawkins: Hey, glad to be here. Thank you
Shayle Kann: All right. My first question for you is, um, my sense is that when people talk about, like, what’s the rate limiter for the growth of data centers, um, you know, there’s a list of things that people point to, the chip supply chain being a big one, power and power availability being another. Maybe right now community pushback being another one that has emerged.
Uh, one of them though– Yeah. One of them, though clearly is, is also labor and the workforce. Um, but I think it gets the least airtime of them. Um, you know, if you’re to think through for you guys, what stops you from, you know, s- skyrocketing your build-out rate, where does labor rank?
Raymond Hawkins: Yeah. Today, uh, you know, you, you listed probably one and– one A and one B power and, uh, uh, is, is the biggie and community has raced up the list pretty quickly. Um, you know, did-didn’t think we’d be at the White House six months ago, um, talk-talking about, uh, who pays for, um, infrastructure on the, on the grid, but we are.
Um, and, and right after that would be labor. Um, and I would tell you, you know, it’s interesting the question of, “Hey, is it stopping us?” Um, what it’s doing is costing a lot more money and putting a strain on the schedule. Lots of, um, lots of management of where we can get the right people slotted in at the right time.
Um, if we had all the labor in the world, we could go faster. Um, but, uh, there, there’s without a doubt, and especially in the electrician space, it’s a tough space to get enough bodies, uh, in enough hours ’cause most of our projects we’re trying to run them, you know, 24/7
Shayle Kann: Yeah. So what does that actually look like? I mean, you alluded to it, but I’m interested in getting a little bit more detail. Like, you know, you- you’re building a new data center. Um, you’re gonna need a certain amount of trained electrician work, a certain number of person hours, let’s say. Like, what is the actual process to go find that workforce and then, you know, presuming that it is constrained, as you’re saying, how do you manage that?
Raymond Hawkins: Yeah. So, so, uh, we, we could probably find, um, people with a better answer than me, but at the end of the day, it starts with our selection of GCs and, and it starts with us building a relationship in the marketplace and going through the bid process early on and going, “Hey, this is what our labor pools are going to look like.
Let’s think about this today. Let’s think about this long before we get there.” Um, there is a lot of moving labor around. Um, we have a large project going in Mississippi. Um, you know, when I tell people we have a large project going just outside Jackson, Mississippi, some people go, “Where’s Jackson, Mississippi?”
So kind of the point of it’s, it’s not, it’s not in downtown, you know, Atlanta or downtown Chicago. Um, so there’s lots of work ahead of time thinking about where we’re gonna find the labor? What markets are we gonna go to get that labor? And is that labor willing to get picked up and moved? And then when we do pick them up and move, is there enough housing for them?
Um, you know, I, I think sometimes, uh, a popular show, “Landman,” you see those, um, in “Landman,” those camps where there’s just trailer and trailer after trailer. There’s a lot of that at data center sites where you are bringing the labor, uh, especially in these tier three, tier four markets, and providing places for them to live and eat and stay so that they can be near the job ’cause, you know, 40 miles outside Jackson, Mississippi, doesn’t have 500, you know, highly trained electricians
Shayle Kann: Yeah. You know, I, I wonder whether, I guess this must be, the answer to this must be yes, but, like, is there a new category of, like, journeyman electrician who is, you know, going and spending six months at a time or whatever it is at living in a camp in outside Jackson, Mississippi, and then in Abilene, Texas, and then in where, you know, wherever the next data center is gonna be?
Is this, is this a whole new category of electrician?
Raymond Hawkins: We have seen, um, teams that will manage a team environment where they’ll roll guys in, you know, 30, 60, 90 days at a time. You, you know, I, I spent four years in the Marine Corps. It feels a little bit like days of deployment where, hey, you know, we’re gonna get ready, we’re gonna go get the ship ready, and it’s gonna go out, and it’s gonna be in the water for six months, and then we’re gonna come back and switch everybody out.
It feels a little bit like that, right? Because the projects are so big and so long that we are bringing waves of talent, waves of skill into the marketplace, giving them a place to live, eat, sleep every day, and then going to the job site where they’re, where you’re bringing, uh, you know, literally an army of electricians and, and to your point, at all different levels.
But yeah, the, the journey part of it or the travel part of it, just like travel nurses, there’s a lot of that. Uh, it’s all over our marketplace
Shayle Kann: You mentioned, especially, electricians. I wanna double-click on that for one second because, you know, it’s not just electricians that you need to build the data center. There’s all sorts of construction that gets done. There’s plumbing, als- other things like that. Is your sense that, um, it’s just that the supply-demand balance of electricians is most acute?
People do say electricians seem to be the one that, like, is the biggest bottleneck. Is that, is that why or is it just, like, a disproportionate number of electricians that you need ’cause the data center is basically a ton of wiring?
Raymond Hawkins: Yeah, yeah. I think it’s a little bit of both, right? Some of it is there’s so many electricians and the level of electricians you need, right? We, you know, the panel that gets put on a data center doesn’t look anything like the panel at my house. Um, so, so, you know, you’ve got much more senior, much more capable electricians, one, and then two, you need hundreds of them, right?
‘Cause to your point, at the end of the day, the reason we charge rent based on kilowatts is it is the limiting factor in the building. And so one, we have really, really skilled electricians, really, really high-end electricians working on really, really, really big systems. It is the biggest piece of the skilled labor.
Yes, there’s a, a, you know, heavy equipment operators and yes, there’s, you know, plumbers and yes, there’s sheetrock people, but all of those look much more like a traditional build. The electricians are much more involved and much more sensitive to the facility
Shayle Kann: That raises kind of an interesting question, which is, um, is the bottleneck predominantly just there are not enough electricians in the country, let’s say? Uh, or is it that there are plenty of electricians, but actually you need this specialized, you know, higher level trained, uh, subset of that category?
And if only you could– So like we’re gonna get to “what do you do about this,” right? But is, is the fundamental thing you do about this get more people trained as electricians or is it get more electricians trained as specialized electricians basically?
Raymond Hawkins: So, so in good AT&T fashion, I think it’s and. Um, I think we need more electricians and we need more senior electricians. It’s both. It is not either/or. Um, we– the– you could employ more electricians today without a doubt, and you could employ more higher-end electricians today. The– just the industry is swallowing up the capacity.
I mean, we– just like you see in the utility provider struggling to keep up with our industry from a pace perspective, same thing with the electrician, for lack of a word, industry or market, has struggled to keep up the pace of the growth. I mean, it’s, it’s hyperbolic. I mean, right? It’s on the news every day, so everyone knows all about it.
But, uh, you know, to me, I think about it just like the utility providers. Those guys, uh, have never heard numbers this big and never heard numbers coming this fast, and that same problem is man- not only is it manifest in the generation, you know, transmission and distribution side of the business, it’s manifest in the who’s-gonna-wire-this on the other end
Shayle Kann: Are there… I mean, you mentioned at the beginning it’s like, it, it’s an expensive problem for you. I was having a conversation not too long ago with, I’ll say, an unnamed, uh, infrastructure person at a hy- at one of the hyperscalers who said, I think in like half joking fashion, “I think we’re gonna be paying electricians like 2 million bucks a year pretty soon.”
Are you seeing bidding wars for talent? Is that kind of where
Raymond Hawkins: Without a doubt. Yeah. The, the electrician, uh, the cost for electricians, you know, I’m, I’m gonna give you a, you know, estimate, has tripled in the time I’ve been in the business
Shayle Kann: Tripled. And
Raymond Hawkins: yeah. In the last decade, yeah
Shayle Kann: So you would think, you know, if you believe in an efficient market, um, that with some, some period of time, the ability to earn triple wages, uh, effectively becoming an electrician would sort of naturally result in a larger workforce.
Maybe not overnight because it takes time to train, but like, you know, even in the absence of, of some of the more, the clever and new programs that folks are, are pursuing, including yourselves, you know, the market would fix itself, you would think.
Um, do you see that happening?
Raymond Hawkins: So I agree. I believe in an efficient market. I believe that markets draw talent when there’s money to be earned there. And, and that is happening. We see it happening in basically every market where we work, where there are people, uh, creating training programs, creating, um, uh, fast tracks, creating methods and means for people to get up to speed and to supply more capabilities in the marketplace because the demand doesn’t– I mean, and that, that’s the other thing.
We talk about a, a, a lease that gets signed today. We’re gonna be building that campus for four or five years. So, you know, being able to, I mean, you could start with someone today and spend two years with them in an apprenticeship, and they’re still gonna be useful on our campus for three or four years while we’re building there.
These things don’t go up overnight. It’s a lot of work for a long period of time. And so that’s one of the reasons we’ve tried to get into the markets where we are and start education programs, because we’re like, “Hey, look, we’re, we’re gonna be here for a while. We know we’re gonna need people at the other end of this.
Let’s start training them ourselves.”
Shayle Kann: Yeah. So tell me more about that, what, what you guys are actually doing. And also, you know, I, uh, one thing I’ve learned about this space is that, you know, it’s a complicated model, or at least historically has been sort of a complicated model for how new, let’s just focus on electricians still, get trained and how they pay for their training, for example, right?
A lot of them will do it through, through federal grants and so on. It’s not historically the case that, like, an employer would say, “I need a ton of new electricians. I’ll pay for their training,” or, “I will sponsor it,” or something like that. So that’s kind of a new phenomenon. So you as the ultimate employer, or maybe not even the employer, right?
‘Cause maybe your GC is the employer.
Raymond Hawkins: Right. All right.
Shayle Kann: You’re just the one who needs them to exist. What is the role that you can play in, in manifesting them?
Raymond Hawkins: Yeah. So, so, you know, we talked about, I, I mentioned that, you know, we’re building campuses and those are multi-year projects. So we’ve looked in the markets and, and we’ve, you know, started the projects and we’ve said, “Hey, um, what do we need to get this project finished? We need more skilled labor. We need more…”
And you’ll hear us use the phrase me- mechanical, electrical, and IT skills and, uh, in, in a marketplace, and can we help the marketplace provide that skill? Um, the biggest campus we have to date is in, uh, just south of Dallas in a suburb called Red Oak, and, uh, we partnered with a, a, a community college there, Texas State Technical College, and said, “Hey, if we were able to convince our partners alongside us to build a curriculum, would you be willing to t- um, s- you know, for lack of a better word, pilot a program where we train in those three disciplines?”
And so in partnership with Texas State’s, uh, Technical College and multiples of our partners, um, ’cause to your point, right, the end user customer is the general contractor or the electrical contractor. He’s the one who needs the employees. Um, our partners, think of Schneider Electric or Siemens, people whose equipment needs to get plugged in by those electricians, have all helped us stand up a, uh, a school at Texas State Technical College, a program to train people.
12 weeks, you start off, you don’t have to know anything about the data center business, and you leave on the other end, uh, a, an entry level but prepared to go work in the data center business on the construction side. Uh, we are– I can’t, I can’t remember if we just completed our second cohort or just started our third, but we already have the first cohort through the program and already employed, um, and employed at dramatically higher wages than the average in the, in where we’re located.
So for us, it’s, “Hey, there’s a problem,” and it’s not only just, “Hey, we need people to help build our buildings,” but how do we serve the communities we live in? And it’s really a way for us to do two things at the same time, uh, help the communities where we’re trying to build, uh, infrastructure, as well as solve a problem for the industry
Shayle Kann: I do wonder, I guess, you know, uh, you’re, you’re building one massive construction project in a given location. It may be that, I mean, if you’re in northern Virginia, there’s lots of these, right? But it may be that at this location outside Dallas or the one outside Jackson, Mississippi, it’s not like there are gonna be a bunch of data centers popping up in that same region.
So is the notion that you’re training people on this specialized set of things, they’re gonna have a few years of employment while you’re building out this one big campus, and then thereafter they, they go travel to wherever the next data center is gonna be?
Raymond Hawkins: I think there’s really two tracks after, after the campus is done. First of all, it’s years, so that Dallas campus will take about five years to build, but to your point, it will come to an end. And for us, I think there’s two tracks after that. There are long-standing jobs on the campus that have mechanical, electrical, or IT infrastructure skills you can deploy staying in your home market.
And then the industry certainly has, to use your phrase earlier, opportunities for journeymen to move to projects all over, right? There’s Wyoming projects and there’s projects, you know, ours in Mississippi. There are lots of more remote, large projects going on, the Panhandle of Texas. There’s lots of big projects, and we certainly expect there’ll be people, um, just like we had to do in Jackson, moving to go work there for four or five years
Shayle Kann: Yeah. There’s obviously, in theory, two things you can do about a labor shortage for, uh, for data center construction. One is, find more people, which is what we’ve been talking about. But of course, the other is, require less people, um, which has more to do with, like, the design of the data center and can you prefab some of it and things like that.
Like, how, how much opportunity do you guys see in, uh, just minimizing the number of labor hours that are totally… that are required in total, maybe particularly for electricians?
Raymond Hawkins: Yeah, I can tell you that that’s one that’s, uh, been part of Compass from the very beginning, so I’m, I’m gonna toot our horn a little bit. Um, we, we viewed, um, building data centers as really a manufacturing process from the very beginning of our business, not a stick-built construction project, but rather an assembly project.
And, and our motivation there, um, I know you’re asking about how do we manage it, does it help with the headcount problem? Our motivation there was really a couple of things: speed, um, if, if, um, higher quality, and a more diverse workforce. If you can, you know, if we talk about putting a project in Jackson, Mississippi, um, almost all of the people that are gonna show up in Jackson and do the work are gonna be men, for lack of a… just the dynamic at home.
Someone’s gotta, you know, manage, uh, the, the, the stability of the household, take kids to school and whatnot. And so generally, it’s more… The traveling journeyman worker is generally a man. And so by doing offsite manufacturing, putting together large components of our data centers offsite, it allows for a safer, more diverse, and more reliable product, right?
Safer for the employees, more diverse because giving somebody, “Hey, you can be here from 8:00 to 5:00. You don’t have to live in a different city. You can be home in the evening with your family.” It allows us to employ more women in the business, and it makes the business much safer to operate, right?
Instead of doing complete stick build in the field, being able to test and commission and assemble offsite and just roll it in on a truck, my boss gets really mad when I say this, and snap it together like Legos. He does not like the Lego analogy, but it really works for me. Um, that is much safer, it’s much more reliable, it’s fewer hours, it’s far less risk to the people on site.
Um, so for us, our motivation was to make the sites safer and faster, but what it has led to is a requirement for less skilled labor. Less skilled labor at the job site. There’s skilled labor now, but it’s now spread around manufacturing facilities with our partners.
Shayle Kann: What can you actually prefab? Like, what… I’m trying to imagine, what are the Lego building blocks?
Raymond Hawkins: Yeah, yeah. Good question. So, um, almost everything in the electrical world and almost everything in the mechanical world are prefab for us now. So think of all the way out from where the utility comes onto the campus and I’ve got a transfer switch or I’ve got a transformer that I’m stepping down, we’ll have all of that preassembled on a, on a skid so that you don’t have to have somebody wire between the transfer switch and, and wire between that and the, uh, you know, a, a, a transformer.
Um, we’ll have the entire power center, everything in there that, uh, d- distributes electricity into the building all show up in one unit. Um, we have in our entire, uh, today both liquid or hybrid cooling, so both liquid and air shows up in a single packaged unit. So these things you’re not having to put together any of it on site.
Um, I… all of our buildings are tilt up, you know, um, prefabricated so that the panels show up and we literally just crane them and set them in place. So no one’s having to make any of the walls, no one’s having to pour any, uh, forms or, or build any forms or pour any concrete. They all show up on a truck, again, literally just like Legos.
Slab A goes in slot A, slab B goes in slot B, and, uh, they get lifted off a truck and slid into place. So almost the ent- uh, the roof is the same way. Uh, we have prefabricated double Ts that support the span. Literally the whole building gets, almost the whole building, it’s about 85% of the building gets snapped into place showing up on the flatbed of a truck
Shayle Kann: That’s, it, it, it’s, it’s remarkable that despite all of that, it’s still a five-year construction project, and also you still need all of these laborers, and you’re still having a short… Like, you’ve done so much to modularize it and prefab it and, and yet, uh, there’s still so much more that needs to be done
Raymond Hawkins: Yeah, and the five years, uh, on the example in Dallas, that’s because it’s 10 buildings, right? So it takes us about nine months to build a single building. And, um, and, and then, and then, um, you know, you go, “Well, wait a minute, Raymond, that doesn’t
Shayle Kann: the scale of a building? Sorry, what’s,
Raymond Hawkins: Um, so yeah, f- f- yeah, 40 megawatts, uh, a little over 200,000 square feet.
Uh, help people understand that, think of, uh, your Sam’s Club. Your Sam’s Club’s probably somewhere between 30 and 40, maybe s- big ones are 50,000 square feet. So think five to six of your Sam clubs under one roof, and lots of electricity. Lots.
Shayle Kann: four, 40 megawatts a piece,
Raymond Hawkins: 40 megawatts is what we build today in that,
Shayle Kann: Okay. So if you’re building five of those, that’s a 200 megawatt data center. Or if you’re building nine of them, I guess you said that would, that’d be a…
Raymond Hawkins: Yeah, the one in Dallas is 10, so it’s 400 megawatts. Um, we, we have lots of fives and lots of tens around the US, so lots of 200s and 400s. And, and in the world of data centers today, you know, I, I remember, uh, when, you know, been in this business 13 years, I remember we talked about kilowatts.
Um, it is completely normal to talk about gigawatt campuses today.
Shayle Kann: Right.
Raymond Hawkins: Yeah.
Shayle Kann: What’s the rate limiter on the building? Or not the rate limiter, what stops you from having a, you know, gigawatt building?
Raymond Hawkins: So, um, for, for us, we, we’ve chosen the form factor we have because we have some beliefs about, um, you know, how you impact the environment, how you impact microclimates around the building. So, we’ve done a lot of studying. Lots of CFD happens inside the building. Hey, computational fluid dynamics, how am I gonna move the energy around?
How am I gonna move the heat up? How high do I move the ceiling? Do I have to spend any money letting the heat move around the room? There, there’s lots of study in what’s going on inside the building. We’ve done a good bit of study on what goes on outside the building, right? When you have all of this equipment, um, not only exhausting but also sucking air in and, and pushing, uh, warm air out, you can produce microclimates.
So for us, we’ve chosen the 40-megawatt, um, form factor to allow us to spread the buildings out and not impact the performance of the equipment. Can you do it in larger form factors? You can, absolutely. Um, do I think that we’ll move our form factor up? Not anytime soon. We felt like we figured out a sweet spot between the equipment we buy, the, the, the readily available nature of the equipment we buy, managing the building that size.
But certainly people that do purpose-built buildings that are very specific engineering can build larger form factors. It’s just not in our roadmap at this time
Shayle Kann: I think we should spend a minute on the community stuff. I mean, it’s tied to the, you know, you mentioned that is one of the reasons to invest in a workforce that’s local and so on. Um, but it’s also obviously become, it’s flared up as like a really big issue now. What is your perspective on what best practices should be for a data center developer entering a community to position yourself to be able to actually build the thing and not get the kind of backlash that we’re seeing right now?
Raymond Hawkins: Yeah. Shayle, I m- I might, I got lots of thoughts on this one, so I might ask you to, uh, tap the guardrails and keep me on message here. But I’d say the biggest one for us is our industry grew up being, um, secretive. Not secretive because we had anything to hide from anybody, but think about the first two significant customers for large scale data centers.
It was the financial institutions and it was the government. And they were secretive about where that large compute footprint sat purely from a security perspective. There was nothing to hide from the community, but you didn’t want anyone to know where JP Morgan’s, you know, bank processing big computer sat.
So nondescript building on the edge of town that nobody knew who it was. So our industry grew up with this secrecy only because it was part of a security plan for very crucial assets, whether financial or government. That’s the history of the business, and we’d never been asked really to be transparent about where we built.
So this notion that somehow we’re, uh, the industry was hiding anything, it was part of the security fabric of the early customers. Um, now, um, that there are data centers going in lots of places and there’s lots of questions, the industry needs to push the envelope to say, “Hey, I need to embrace the communities that I’m in, and I need to let people know what I’m doing,” because, you know, and I’m gonna give a weird analogy, right?
People were not crazy about trains when they started. People thought that if you traveled 30 miles an hour, it would cause your heart to stop, and people said, “Hey, I don’t want trains in my backyard.” Um, Edison had to give away the light bulb. No one wanted to buy it. And we can’t imagine that today, right?
But people had to be… You know, he had to give a light bulb to someone to get them to use it. And I’m not, and I’m not comparing us necessarily to trains or light bulbs as far as transformative, but digital infrastructure’s here. It’s not going away. All of us want our cell phones. And what I think, you know, in, in the world of Compass, we think about a Class A developer.
What’s a Class A developer look like? He embraces the place where he’s going. He embraces his community and says, “Hey, I want to be a good citizen in the community. I happen to be a corporation, and I happen to run a building,” but no different than being a port or a harbor or a train or any other infra– a dam, a water treatment facility.
“I’m gonna be part of the infrastructure, and I want to be a good community citizen in that infrastructure.” Um, being, um, transparent about where you go, um, embracing your community, uh, offering to help in your community, being a good corporate citizen from a support, uh, you know, and, and, you know, paying, um, for the things that you’re gonna use, right?
Um, to us, that’s all part of being a Class A developer. We use that term internally to say, “Hey, how would you…” You know, in the commercial space, people think of Class A office space, and it means something. And we think the same thing about the data center space. What’s a Class A data center developer?
They’re open with their community. They’re helping their community. They’re a good s- corporate citizen in the community, um, helping not just from an infrastructure standpoint, but helping from a, a education, training, jobs, all of those things
Shayle Kann: That makes sense to me. I think speaking of education, I guess the other sense that I have is that, um, there are some specific concerns that communities have about data centers, and then there’s some amorphous ones that are not really localized but are, but are a thing. Like, the specific concerns, you know, that you hear about are w- things like water consumption, um, impact on electricity bills, things like that.
And then there’s some amorphous ones that are like, “This is bad for my community, and I don’t totally understand why I think it’s bad for my community, but it’s, but it’s bad.” And, and I feel like there’s a big education challenge, right? You have to convince everybody that their rates are not gonna go up, that the water consumption’s not gonna be an issue, that the noise won’t be an issue, whatever it is, the specific things.
And then you also just gotta have the– y- figure out a way to give this generalized sense of this is actually good for us and not bad for us. And I, I, I wonder how you approach the education, community education stuff.
Raymond Hawkins: So, uh, I think you nailed three of the big ones. Um, so we get asked about water. Um, you know, I talked about, um, you know, being transparent, you know, secrecy. Uh, you, you, you mentioned infrastructure, the cost of electrical infrastructure. Um, the other one is, you know, do, do I need this? Is this generally this bad for me?
What’s that big building over there that looks, you know, weird? Do I need this in my community? Um- Let, let’s go through them real quick. Water. So I’m gonna do a quick Compass commercial. We’ve been waterless, uh, cooling since the day we started. Um, you know, the industry with GPUs has forced, uh, you know, a higher, a, you know, a more dynamic, uh, heat rejection, um, model, so liquid cooling comes into play.
We just do it with a closed loop, so we’re not a constant user of, um, utility water in the marketplace. We think that’s important. Um, frankly, most of the industry is going that way, but it’s completely fair for people to ask the water question. I just think you need to talk with your developer, talk with the data center in your space, and make sure that the water they’re using is not coming out of your utility and it’s closed loop.
In that regard, I think you largely handle the water conversation. Um, paying for the electrical infrastructure. For us, um, we’ve taken it on ourselves to pay for the infrastructure that delivers power to us, right? There’s generation that’s owned by, you know, generally the utility, then there’s transmission and distribution, um, and then the substations to get your site powered.
Um, we’ve taken the position from the very beginning that we pay for anything downstream to get to us, um, and, and put that on our bill so that it doesn’t impact ratepayers. Not– And then we take that one step further. We also wanna be a resource to the marketplace. Um, I saw a video you did about your, um, electric car sitting in your house and how much, uh, storage was in that facility in the car.
Uh, I think you compared it to, um, uh, wall units from Tesla. Same thing, right? I build a data center, I got tons of generation on site. Me and my customer paid for that generation. Not only do we wanna pay to get the electricity to our campus, but in the worst times, we wanna be a resource to the grid. We wanna be able to generate, uh, or, you know, or turn off our need for grid power and generate it ourselves.
So we try to collaborate in the markets and say, “Hey, how can I be a resource for the grid in your most strained times?” Whether it’s Snowmageddon in Texas or whether it’s summer July in Phoenix, uh, we wanna be able to say to the, to the… As a Class A developer, we wanna be able to say to the utility provider as well as the customers around, “Hey, not only are we gonna pay our own way, we also wanna be able to help the grid in the toughest times.”
So that’s water, that’s electricity, and then, “Hey, I’m just not sure data centers are good for me.” Um, it’s really a big education, right? You know, w- w- we, uh, I’ve– This is not my phrase. Uh, I read it on an article, uh, yesterday, “The analog American,” the American that doesn’t realize they’re, they’re reading about data centers on their phone and they’re tweeting about it on their phone and saying, “I don’t want it,” and all of that happened in a data center somewhere.
Um, all, all of the things that you’re reading and all of the things you’re tweeting and saying you disagree with all took place in a building somewhere called a data center. And, and not, not to be ugly about it, but just to say, “Hey, realize we’re, we’re not going backwards. We’re not gonna get away from a digitally connected world, and how do we partner with that digital infrastructure in such a way, how do we partner with our communities in such a way that doesn’t scare people?”
Hey, this is a good thing for you. Um, it’s a good thing for, for the, for the business. It’s a good thing for your community, and can we do it in a way that makes everybody feel comfortable and safe?
Shayle Kann: Raymond, this was, uh, a lot of fun, and I, I, I will say I really appreciate the way that you frame this, uh, both from the labor perspective and the community perspective. I think, uh, I think there’s a lot– It’s gonna be… It’s a tough road for data center developers right now, just, and, uh, from both the practical perspective of getting enough people, and then from the, you know, perspective of getting enough buy-in and, and license to operate.
So there’s a lot of learning to be done there, so I appreciate you thinking hard about it.
Raymond Hawkins: Yeah, absolutely. It’s important for us, important for us as a business, but certainly important for us. I mean, we could take it all the way to, you know, um, national security, right? I think, um, having a robust digital infrastructure, I think makes us a better country, a safer country, and, uh, it’s, it– I think you can’t under- you can’t overstate how important it is.
Shayle Kann: Yeah. All right, Raymond, thank you for your time
Raymond Hawkins: It’s been great. Thank you so much
Shayle Kann: Raymond Hawkins is the Chief Customer Officer at Compass Datacenters. This show is a production of Latitude Media. You can head over to latitudemedia.com for links to today’s topics.
This episode is produced by Max Savage-Levinson. Mixing and theme song by Sean Marquand. Ann Bailey edits the video version of the show. Stephen Lacy is our executive editor. All of our episodes are now on YouTube. Subscribe to Latitude Media for episodes of this podcast and Open Circuit. You can find the audio version of the show anywhere you get your podcasts.
I’m Shayle Kann, and this is “Catalyst.”
