WEBVTT
00:00:00.160 --> 00:00:06.080
The data centre industry is moving faster than ever, and behind the scenes some seriously smart people are making it all happen.
00:00:06.240 --> 00:00:13.039
Welcome to the Inside Data Centre Podcast, where we dive into the stories, insights and innovations driving real change across the industry.
00:00:13.279 --> 00:00:19.920
From hyperscale campuses to game-changing technologies, we're talking to the people shaping the future of digital infrastructure.
00:00:20.079 --> 00:00:25.839
Whether you live in breathed data centres or just want to understand what's powering our connected world, you're in the right place.
00:00:26.079 --> 00:00:27.920
Welcome to the Inside Data Centre Podcast.
00:00:28.000 --> 00:00:33.840
On this episode, I'm joined by Tom Babbington, founder and CEO of Adaptive MDC.
00:00:34.159 --> 00:00:35.200
Good afternoon, Tom.
00:00:35.439 --> 00:00:37.039
Hey Ali, how are you going, mate?
00:00:37.359 --> 00:00:37.840
I'm good.
00:00:38.000 --> 00:00:39.039
Second time on the podcast.
00:00:39.119 --> 00:00:44.000
I had to look back before we uh started this conversation to see when it was, two years ago, 2024.
00:00:44.240 --> 00:00:46.640
But that time we were talking about a very different subject.
00:00:46.799 --> 00:00:49.679
We were talking about Data Centre Cycling Club.
00:00:49.840 --> 00:00:54.240
So I think uh we're we're in a slightly different area remit now for this conversation.
00:00:54.560 --> 00:01:00.159
Yeah, I I happily uh talk about cycling, but it it it it might not get the same uh same traction.
00:01:00.399 --> 00:01:01.439
Yeah, it was.
00:01:01.920 --> 00:01:02.560
Yeah.
00:01:03.039 --> 00:01:05.840
And I think um data center cycling club's still going as well, obviously.
00:01:05.920 --> 00:01:06.719
We're still good?
00:01:06.879 --> 00:01:09.120
Yeah, so we did we did the ride in CAN this year.
00:01:09.280 --> 00:01:13.120
Um we were less ambitious, more inclusive, which has been the feedback.
00:01:13.200 --> 00:01:20.319
So we did I think the the the we did four days and we did 150, 120, 150k average each day.
00:01:20.400 --> 00:01:23.280
So there's 20 riders, so it was bigger.
00:01:23.519 --> 00:01:29.920
Um we raised um plenty of money, which which goes to Action Medical.
00:01:30.000 --> 00:01:31.920
Um yeah, it was it was a good crack.
00:01:32.239 --> 00:01:32.879
Yeah, well done.
00:01:32.959 --> 00:01:36.319
And obviously, if anyone wants to get involved in the future, I'm sure they'll be able to find it.
00:01:36.400 --> 00:01:39.760
Data center cycling club has a website and everything, so it should be easy to find.
00:01:39.840 --> 00:01:41.040
But we'll put it in the show notes as well.
00:01:41.120 --> 00:01:41.680
You know, you never know.
00:01:44.480 --> 00:01:46.560
There's always some, always some lurking.
00:01:46.719 --> 00:01:48.480
You can't you can't hide for them.
00:01:48.799 --> 00:01:49.519
Definitely.
00:01:49.680 --> 00:01:56.079
And back to the the world of um work and and data centers, not that cycling is not work, it's hard work, not that I cycle.
00:01:56.159 --> 00:01:59.760
But um, you've recently launched Adaptive MDC.
00:01:59.840 --> 00:02:05.439
So, say we're in a very different position to where we were a couple of years ago when we had a conversation from a from a work perspective.
00:02:05.599 --> 00:02:10.080
Do you want to give a bit of an overview and just explain how you've arrived at this point?
00:02:10.400 --> 00:02:11.919
Yeah, um absolutely, thank you.
00:02:12.000 --> 00:02:22.159
Yeah, so so adaptive uh MDC is uh a modulat business where we take the traditional data center problem and turn them into modular solutions.
00:02:22.240 --> 00:02:26.879
So everything from design, manufacturing, construction, integration through to commissioning.
00:02:26.960 --> 00:02:33.360
Um, so like the full life cycle um where we try and drive through efficiency throughout the whole process.
00:02:33.520 --> 00:02:43.520
Um lots of things, so uh you know, AI has changed, the density sort of uh the the these facilities having smaller footprints lends itself to modular.
00:02:43.599 --> 00:02:54.400
Um, and we're we're looking sort of just just improving sort of the execution of the data center delivery um or optimizing it as best we can.
00:02:55.120 --> 00:02:59.039
And your background in the industry, how did it kind of lead you to this point?
00:02:59.120 --> 00:03:02.400
Because I guess I think a lot of the time we learn from experience, don't we?
00:03:02.560 --> 00:03:08.159
We see these challenges, and then some people create a solution which becomes an organization, which is obviously what you've done.
00:03:08.240 --> 00:03:13.039
But what how is your experience sort of in the world of data centers led you to this point?
00:03:13.680 --> 00:03:23.039
Uh so I I've been thinking about this a lot as sort of my route to the data center market is probably quite traditional in that I never planned on it happening.
00:03:23.280 --> 00:03:25.199
Um, I just sort of evolved into it.
00:03:25.360 --> 00:03:33.759
My construction career started in Afghanistan um in 2006, and we were building uh facilities for the vulnerable in the community.
00:03:33.919 --> 00:03:38.400
Um then I came back and had my one and only construction project in the UK.
00:03:38.560 --> 00:03:48.719
So six years then six years after that, I I'm on the hyperscale uh job in Finland um looking after uh uh elements of delivery for that.
00:03:48.800 --> 00:03:56.639
So I've I've spent the majority of my career working overseas, I think about 42 countries and counting in in different locations that I've worked.
00:03:56.960 --> 00:04:11.360
Um I am happiest when I'm in a room of of people from different cultures trying to understand what their problems are, how to solve them, um, rather than just sort of sticking to what the book says, the traditions of this is how it's done.
00:04:11.439 --> 00:04:13.520
It's like the the world isn't made up of that.
00:04:13.759 --> 00:04:21.839
It's made up of of problems that that need to be solved, and and and ultimately people people create them, so you've got to understand what it is that they they need solving.
00:04:22.319 --> 00:04:31.600
Yeah, and I think from the culture perspective, it's quite an interesting point as well, because this industry is full of different cultures, diverse, you know, like different people on these projects.
00:04:31.839 --> 00:04:37.360
Is that you'll say you've got probably more experience than most working and operating in that environment?
00:04:37.600 --> 00:04:46.639
But do you think that it's important for people to be successful in this industry if they have that ability to kind of flex and mould to all these different cultures?
00:04:47.040 --> 00:04:52.639
Well, I think uh inevitably the greatest thing about our industry is the people that are in it, right?
00:04:52.720 --> 00:04:56.560
You know, that's that's how it's been able to grow and be so successful around the innovation.
00:04:56.720 --> 00:05:00.800
But where we're struggling now is finding enough good people to execute.
00:05:01.120 --> 00:05:06.319
So, you know, for years we've talked about a resourcing problem, there's a capacity problem.
00:05:06.480 --> 00:05:14.560
So, yeah, you've got to you've got to sort of see what opportunities lay ahead in terms of the people and the capacity and what they can do and find out how they can solve it.
00:05:14.720 --> 00:05:25.920
So, yeah, um, you know, there's there's been times where I've been in we had a I was working on a hotel uh construction project in Tunisia and we couldn't get the guys to uh work any later.
00:05:26.079 --> 00:05:33.199
So we were like paying them in advance and getting them to work later because the the job was running late and they would take the money and not turn up for another week.
00:05:33.360 --> 00:05:35.759
So we decided not to pay them, then they didn't turn up.
00:05:35.920 --> 00:05:38.319
So we just said, Well, what are they missing?
00:05:38.399 --> 00:05:39.759
We we cook them dinner.
00:05:39.839 --> 00:05:40.879
So it's simple as that.
00:05:41.040 --> 00:05:42.959
Every night there was dinner for all the staff.
00:05:43.040 --> 00:05:54.560
They ended up bringing all their families, but they stayed until the the day was done, and it it improved productivity, and and it's a it's a it's a small part of of how we how we do it.
00:05:54.720 --> 00:06:07.120
Um so I think this industry has to has to find a different way rather than just going out and recruiting straight from university or straight from primary school now, because it's like you've got to get the good people in, you've got to find a different way.
00:06:07.680 --> 00:06:09.279
Yeah, 100% interesting.
00:06:09.439 --> 00:06:18.959
And on the modular point, which is obviously why we're here, I think it feels like I've been talking about modular data centers for the whole time of since I started this podcast, which is coming up to six years.
00:06:19.199 --> 00:06:32.879
But I think now the the conversations are becoming more kind of detailed and and people it's a real focus now for companies, and it's a solution that the industry has to you utilize in order to kind of deliver at the scale that that it is.
00:06:33.040 --> 00:06:41.839
But you've do you think sort of from your perspective that now is the time where the industry's really pushing towards this new sort of this model of working?
00:06:42.319 --> 00:06:45.680
Uh yes, um, of course.
00:06:45.759 --> 00:06:51.199
Uh, but I don't think fundamentally anything's really changed in the industry.
00:06:51.360 --> 00:06:58.079
I think the demands are still high, it's still great, like there's still an expectation to deliver something in a in as quickly as possible.
00:06:58.319 --> 00:07:04.480
I think it's the biggest change is the amount of people asking for it to be delivered in this unachievable time.
00:07:04.639 --> 00:07:13.759
So customers have always said we need this online in 12 months, but now it's the number of customers is exponential, so we have to find alternatives.
00:07:13.839 --> 00:07:23.920
So and you know, innovation has always happened, and we're an industry full of great engineers who are coming up with solutions to solve these problems.
00:07:24.240 --> 00:07:29.759
It's finding the um the way of putting them into practice.
00:07:30.000 --> 00:07:31.680
I keep saying it's like the execution.
00:07:31.839 --> 00:07:41.600
How can we how can we standardize a certain design to be able to solve to to to deliver that?
00:07:41.759 --> 00:07:44.560
Um and and that's where modular lends itself.
00:07:44.800 --> 00:07:56.240
Plenty of companies are doing different elements of modular, whether it's their their sort of the productizing, their components, and it's bringing them together to make sure you can just control as much as possible.
00:07:56.480 --> 00:08:02.480
Now I'm not saying it's standard, you have to design a solution to meet the customer's requirements.
00:08:02.560 --> 00:08:03.439
That's not going to change.
00:08:03.680 --> 00:08:21.120
They will never buy a data center just off the shelf because the nuance of what they want to go in it or how they want it to operate will change, and and that's part of what what our approach is is making sure the design is is sort of scope-specific, but optimising what's already available.
00:08:21.759 --> 00:08:32.639
You touched on innovation as well, and I I guess what we've seen a lot in the industry over the last um say two years is a lot of new innovative organizations enter the industry.
00:08:32.799 --> 00:08:44.159
And I think what that's brought is a different perspective and looking for different ways of working rather than the legacy data center market that kind of had a way of doing things and that was just how how we did things.
00:08:44.320 --> 00:08:50.000
Do you think that is also impacting on execution or how we build and how we deliver data centers?
00:08:50.720 --> 00:08:54.639
Well, yeah, I think I think they both they both there is a place for both, right?
00:08:54.720 --> 00:08:56.399
So it's it's quite a broad market.
00:08:56.559 --> 00:09:06.159
There's you know the the location of the data centers will dictate the sort of how they're gonna build it, but the speed and scale is is the alternative.
00:09:06.320 --> 00:09:17.360
So um some of the biggest jobs, construction jobs happening at the moment are going to be modular in in the Middle East, and that's just because they can control so much so many elements of it.
00:09:17.519 --> 00:09:26.000
But the smaller market where concrete is king and they've got workforces in place, of course, they can go and and use the traditional method.
00:09:26.320 --> 00:09:34.480
So it doesn't solve every problem, it just provides an alternative that allows us to be to be much more efficient with the resources that are available.
00:09:34.639 --> 00:09:56.480
And there is there is plenty of manufacturing capabilities across the across the world that can help provide modular solutions, is just having the expertise around the the data center side of it to make sure the quality and the standardizations within it are suitable for our very demanding but uh but lovely customers.
00:09:56.960 --> 00:09:58.080
Very lovely customers.
00:09:58.240 --> 00:10:00.720
And has geo geography impacted as well?
00:10:00.879 --> 00:10:17.279
Because again, I if I if I think back over the conversations I've had over the years, a lot of the conversation when we were talking about modular or similar type solutions was the ability to build in more remote regions where there, like you said, there isn't a workforce or there isn't the access that they need.
00:10:17.360 --> 00:10:21.919
Is that again, is that impacted the the sort of the move further towards modular?
00:10:22.240 --> 00:10:23.600
Yeah, absolutely, right.
00:10:23.759 --> 00:10:38.799
Uh it it you know, you you want to get a thousand people to work up in the north of Finland, uh, or you build it in a in a controlled environment somewhere else in Europe and ship it to play so you're only on the ground for for a short period of time.
00:10:38.960 --> 00:10:46.000
You can get testing done, or majority of the testing done in a facility off site and then shift it to site so there's less time on the ground.
00:10:46.159 --> 00:10:52.720
It controls health and safety aspects, it controls um the movement of materials in and out of of the same area.
00:10:52.960 --> 00:11:03.039
So it's not it's not the same as um you know we've seen in the past of pick a random location where land and power is cheap and try and get a thousand people over there for 18 months.
00:11:03.519 --> 00:11:06.799
That that's that's gonna have to change.
00:11:07.440 --> 00:11:11.600
Yeah, they are there are still some trying it though, I will I will say that.
00:11:11.840 --> 00:11:18.720
Um from a and and customer, so when you're in your conversations with customers, again, is this is is it now this is a solution they're looking for?
00:11:18.879 --> 00:11:22.799
This is not a solution that they may you know proceed with at some point.
00:11:22.879 --> 00:11:28.639
This is something they need now in order to achieve their sort of unbelievable delivery schedules.
00:11:28.879 --> 00:11:33.759
Yeah, look, it's it's where we're spending a lot of our time is developing reference designs for customers.
00:11:33.919 --> 00:11:48.320
So they have a standard go-to-market solution where depending on the location and depending on slight variances in terms of what what what deployment they want, we can we can change our design.
00:11:48.480 --> 00:11:58.159
So reference designs from everything from five kilowatt racks up to 130 at the moment, but can go up all the way to 300, 600 when that comes about.
00:11:58.320 --> 00:12:16.480
It's about configuring the design to be fully modular, so it's not just a kit of parts, it's it's looking at it from a construction commissioning standpoint and how you bring everything in together so your your shipping is as efficient as possible, um, and not just not just boxes of air and putting everything in.
00:12:17.840 --> 00:12:21.360
What scale can we are we talking from a modular perspective now?
00:12:21.440 --> 00:12:36.639
Because again, I I I think the perception to the outside world, or I might it might just be people like myself, the uneducated, that you know, modular they assume smaller, you know, like our smaller facilities, you know, let's say we can do a one megawatt haul and a modular perspective build.
00:12:36.720 --> 00:12:41.600
What scale are we talking now when you're when you're working with these customers for their designs?
00:12:41.840 --> 00:12:46.799
So uh uh limitless uh is it because it is.
00:12:46.960 --> 00:12:49.120
Um how big do you want to go?
00:12:49.200 --> 00:12:52.480
It it we of course there's the one megawatt.
00:12:52.639 --> 00:12:56.480
There's the containerized solution, which is a data center in a container.
00:12:56.639 --> 00:13:05.120
Um, our solution is a modular approach where it's a series of containers that make up a sort of a modular um solution.
00:13:05.279 --> 00:13:12.000
So there's plenty of jobs over 200 megawatts that are modular at the moment that that I know of.
00:13:12.240 --> 00:13:17.919
Um some of the sizes that we're working on are over 100 megawatts.
00:13:18.000 --> 00:13:21.440
Um, but then there is a repeatable two and a half, a repeatable one.
00:13:21.759 --> 00:13:26.000
Um there's a deployment in a in a powered shell of different configurations.
00:13:26.159 --> 00:13:29.039
So it is truly it's limitless.
00:13:29.600 --> 00:13:41.279
Yeah, and I and I think that again is it's sort of busting that perception that I've I always think people think modular is small, like it's strange, it's just the word, I think the terminology leads people to that to that route.
00:13:41.360 --> 00:13:49.519
So again, as as the industry scales, it needs these solutions in order to scale because it can't keep up with the demand and deliver quick enough.
00:13:49.840 --> 00:14:00.000
And we we so we always open our our introductory presentation with uh this is what module data center means to us, and that is it's not containerized.
00:14:00.960 --> 00:14:06.320
There was an example of uh a containerized data center in the footsteps of of the conference at Cannes year.
00:14:06.480 --> 00:14:26.080
That for us is containerized, not modularised, that's just a container that that sir solves a problem, but scaling that to a level where you're trying to get uh you know a huge sort of a gigawatt deployment or 250 megawatt deployment, you have to do a design that is more efficient because that would not be.
00:14:26.960 --> 00:14:27.840
Yeah, definitely.
00:14:28.000 --> 00:14:36.559
And I think if and if you talk about CAN, I think a lot of the outside of the uh the drinks and everything, a lot of the the conversations were around execution.
00:14:36.639 --> 00:14:41.519
And I I genuinely think execution's now become one of the key challenges in this market.
00:14:41.679 --> 00:14:56.879
I think execution and community, you know, we've had power and people for years, but I really think now the conversations I'm having anyway, with with whether it's with funds, investors, CAs, whatever they might be, is around execution and also the community and managing that challenge.
00:14:56.960 --> 00:15:03.200
But do you think the industry is doing enough to really sort of solve this execution problem?
00:15:04.240 --> 00:15:09.200
Um uh no, probably not, uh if I'm honest.
00:15:09.360 --> 00:15:24.639
I think uh talking about it a lot, um, but we all get very excited about the next shiny object, the next big project that's gonna take up all the the power in the land and how we're gonna deliver it um is is sort of secondary to it.
00:15:24.799 --> 00:15:27.120
So I think I think we can certainly do much more.
00:15:27.200 --> 00:15:37.600
I think there can be plenty of collaboration within the industry around how we how we combine forces to to to make sure that execution is paramount.
00:15:37.840 --> 00:15:43.039
Um everyone if everyone stays in their lane, then then nothing changes, right?
00:15:43.120 --> 00:15:51.200
That the execution doesn't change, everyone's just trying to sell their their product because that's obviously how their businesses are stacked up.
00:15:51.360 --> 00:16:06.559
But but you combine that and you bring that all together and you're sort of agnostic in your delivery, or you um have a design that integrates different components from different people because they have capacity online or they have sorry capacity within their their production lines when it's needed.
00:16:06.799 --> 00:16:14.879
So rather than sticking to one that's saying it's 52 weeks, you can't have a generator of 52 weeks, or your switch gear is you know it's it's 18 months.
00:16:14.960 --> 00:16:24.720
It's like, okay, well, if we can now get a customer the same specification in nine months, why would we not do that to get their GPU online?
00:16:24.879 --> 00:16:30.720
Like it still it still serves the same same uh or solves the same problem.
00:16:30.879 --> 00:16:36.720
And I think we we have to um we have to sort of look beyond the next three years.
00:16:36.879 --> 00:16:38.799
Capacity's tied up for the next three years, you know.
00:16:38.879 --> 00:16:40.159
Look at production lines.
00:16:40.320 --> 00:16:48.000
You you you you can't now go to uh many vendors and say, what's your timeline without them saying it's a year.
00:16:48.240 --> 00:16:52.480
How do you how do you get certainty around a any timeline with a vendor?
00:16:52.559 --> 00:16:54.399
It's you go to them with an order.
00:16:54.639 --> 00:16:55.759
That's the only way.
00:16:55.919 --> 00:17:04.160
Um or you've got a deep enough pocket that you can just pay over the odds, which which isn't healthy, which isn't which isn't healthy for the market.
00:17:04.319 --> 00:17:14.400
So I think if you if you go with that collaborative mindset of of trying to solve the industry problem, you'll solve your customer's problem.
00:17:14.960 --> 00:17:15.680
Yeah, and I agree.
00:17:15.839 --> 00:17:19.920
And you have seen, I guess, more flexibility in customer supply chain, haven't you?
00:17:20.000 --> 00:17:33.359
You've seen them have to increase their supply chain in certain areas, not not broadly across everything, but in certain areas they've had to increase their supply chain in order to achieve part of their their targets.
00:17:33.519 --> 00:17:39.119
They're still not there where they need to be, but there is a bit more flexibility now than there was going back two or three years.
00:17:39.359 --> 00:17:40.240
Yeah, I think so.
00:17:40.799 --> 00:17:42.559
There's less rigidity for sure.
00:17:42.640 --> 00:17:50.160
Uh look, there's plenty more alternatives, and you know, the the quality's in general, the quality is improving.
00:17:50.400 --> 00:17:57.359
Um so yeah, I I think I think that's been a real positive is is just the breadth of what's available.
00:17:57.599 --> 00:18:06.319
Um but still it still needs to be delivered uh and and and uh sort of the the execution still needs to be focused on.
00:18:06.960 --> 00:18:07.680
Yeah, definitely.
00:18:07.920 --> 00:18:15.039
And when you look to the future, which is it always seems impossible in this industry because the future seems to arrive pretty much every day.
00:18:15.200 --> 00:18:21.279
Um what what does a data centre look like in you know, one, two, five, pick a pick a number of years?
00:18:21.440 --> 00:18:24.880
Um I think again, like we've seen that evolution, haven't we?
00:18:25.039 --> 00:18:32.799
And data centres are very different uh buildings or facilities to what they were, but you how do you see it progressing over the next few years?
00:18:33.680 --> 00:18:39.279
Yeah, uh so I I thought all data centers are going to go to space, isn't that what Space SpaceX is hoping for?
00:18:39.440 --> 00:18:47.920
Um I think I I read I read most weeks that they are going to space, and then two articles later on my search, it says they're not going to space.
00:18:48.000 --> 00:18:50.240
So I think we're still a bit unknown on that one.
00:18:50.400 --> 00:18:51.039
Depends who you are.
00:18:51.119 --> 00:18:54.960
So what depends if you own if you own a space company, then you say they are.
00:18:55.599 --> 00:19:00.480
Yeah, so so once you've got your telescope out, you can see what the space, the data sensor of the future look like.
00:19:00.640 --> 00:19:04.799
No, look, I I uh it's it's it's an impossible question.
00:19:04.960 --> 00:19:15.119
I think you know about 100 kilowatt racks happening now, 150 are happening now, 600 are just around the corner, um direct to chip cooling is gonna happen.
00:19:15.279 --> 00:19:17.279
All the technologies there, the innovations happening.
00:19:17.359 --> 00:19:25.039
I think what we'll see differently in the future is um the public perception, or I hope.
00:19:25.119 --> 00:19:28.640
I but uh you know the public perception of data centers is gonna have to change, right?
00:19:28.799 --> 00:19:33.680
They're gonna have data centers are gonna have to become part of economic ecosystems.
00:19:34.000 --> 00:19:34.799
It just has to happen.
00:19:34.880 --> 00:19:43.359
It's not they're not a drain on power, they're not a drain on water, not a drain on all the the resources, taking it away from municipalities and and and the people.
00:19:43.680 --> 00:19:45.279
They will will be part of those.
00:19:45.440 --> 00:19:53.359
It allows people to work from home, allows them to work more efficiently, it it gives jobs, it adds uh value to to those environments.
00:19:53.599 --> 00:19:55.039
And I think that'll be the change.
00:19:55.119 --> 00:20:03.039
So what they'll look like from the outside, I don't know, but hopefully people's perception of what they look like and what they are will shift massively.
00:20:03.599 --> 00:20:05.359
Yeah, and I think it's a really important point.
00:20:05.440 --> 00:20:13.759
And and I guess in your career, you you've been in the industry for a let's say for a while compared to a lot of people, even though it's not that long.
00:20:14.240 --> 00:20:28.880
The the the perception of the industry, kind of what I feel is that we've gone from no one really knowing, no one really caring, to all of a sudden it feels like everybody really caring for different reasons.
00:20:28.960 --> 00:20:34.880
It's not related to a lot of it's related to energy or land or what and I completely understand that that viewpoint as well.
00:20:34.960 --> 00:20:36.480
But have you do you feel the same?
00:20:36.640 --> 00:20:41.200
Do you feel like there's just been this sudden shift from not in the news to front page news?
00:20:41.519 --> 00:20:56.319
Yeah, uh I just so yesterday I was talking to to uh a friend in our office and just talking through sort of a an investor deck around uh a funds development for day centers.
00:20:56.400 --> 00:21:03.839
And I was talking through just just basic rule of thumb metrics for day center construction, and he was like, How expensive?
00:21:04.000 --> 00:21:04.480
How much?
00:21:04.640 --> 00:21:14.400
I'm like, yeah, but and I explained why it was that much, explain why the and it's become it's come every day people asking questions about day centers.
00:21:14.640 --> 00:21:23.599
And that was you know, we we met people um actually it was in Cannes, they came over, they have a they have a power station and they're like, I've got power, I've got land, how do I build a day center?
00:21:23.680 --> 00:21:24.799
And I'm like, great.
00:21:24.960 --> 00:21:27.279
You know, it's become ever present in everyone's mind.
00:21:27.359 --> 00:21:44.079
It sort of it's not I I I I think that's that's the biggest shift is it's no longer just explaining that they are the power stations of the future in in that sense, it's it's driving economic change that that has been ignored so far.
00:21:44.240 --> 00:21:46.480
Um so yeah.
00:21:53.440 --> 00:21:59.759
And you end up doing their job by finding the off taker at the end because you you know, fifteen years of experience.
00:22:00.079 --> 00:22:02.319
In the sector, you get to know a lot of people.
00:22:02.640 --> 00:22:06.319
Um, so I mean, I'm guessing you're pretty much the same space.
00:22:06.400 --> 00:22:11.279
It's like how many people come to you and say, 'I've got some power and some land, Andy.' Who do I need to talk to?
00:22:11.759 --> 00:22:12.559
Exactly that.
00:22:12.640 --> 00:22:16.079
Yeah, it's um farmers very often in the UK.
00:22:16.400 --> 00:22:17.839
I speak to a lot of farmers.
00:22:17.920 --> 00:22:19.279
I had a lot of messages yesterday.
00:22:19.359 --> 00:22:30.559
I posted about the new the new data centre in Devon, and I had a lot of messages from residents of Devon asking me why I'm building my data centre sat down there, and I had to explain it's not my data centre.
00:22:31.200 --> 00:22:32.799
I'll just I'll just sharing the news.
00:22:32.880 --> 00:22:33.759
That's that's all I had.
00:22:33.839 --> 00:22:35.200
But yeah, no, it's fascinating, though.
00:22:35.279 --> 00:22:40.640
I do agree, and I think it's all nice to educate, it's it's totally our our challenge to solve solve the perception piece.
00:22:40.720 --> 00:22:49.839
But it's um it's a very, very interesting journey to go from, as I say, like I felt I had to kind of tell everybody about what I did and really explain it.
00:22:50.079 --> 00:22:54.480
You get to a point where you're like, oh, I can't, yeah, I just can't be bothered to go into this detail anymore.
00:22:54.640 --> 00:22:55.599
You don't really care, do you?
00:22:55.680 --> 00:22:58.160
To now everyone's saying, you work in data centres, don't you?
00:22:58.240 --> 00:22:59.200
Like, what what do you do?
00:22:59.279 --> 00:22:59.839
What what are they?
00:22:59.920 --> 00:23:01.200
They're gonna build one near our house.
00:23:01.279 --> 00:23:04.720
You know, it's it's fascinating, but a lot of work to do for the industry.
00:23:04.880 --> 00:23:10.079
Um before we close up, just a couple of quick fire questions, which I ask everyone on the pod.
00:23:10.480 --> 00:23:15.599
If you could ask everyone in the sector to start or stop doing one thing, what would it be?
00:23:16.480 --> 00:23:18.160
Oh, stop doing something.
00:23:18.400 --> 00:23:22.720
Uh trying to think of what people do that that bothers me.
00:23:22.960 --> 00:23:30.079
No, I think I think uh uh I think just stop accepting that that's how it's always been done, right?
00:23:30.160 --> 00:23:33.200
We're trying to move at a speed that we have to find alternatives.
00:23:33.359 --> 00:23:40.240
So uh you know that goes to what I was saying about collaborating, you know, across disciplines and companies and borders.
00:23:40.400 --> 00:23:47.839
I think there's much more we can do if you just if you just sort of stop accepting sort of the normal way of doing things.
00:23:48.079 --> 00:23:51.680
Uh is that a good enough answer?
00:23:52.079 --> 00:23:52.880
That is a good answer.
00:23:53.200 --> 00:23:54.000
I'd be more brutal.
00:23:55.359 --> 00:23:55.839
Yeah, exactly.
00:23:55.920 --> 00:23:56.400
It's different.
00:23:56.480 --> 00:23:57.119
Where's the line?
00:23:57.200 --> 00:24:02.640
But it's uh no, it's a fascinating, it's an interesting point as well, because I feel like the industry is going through that shift now.
00:24:02.720 --> 00:24:08.720
I I do I remember when I first joined, well not joined, but first started where it's not as I make it out like it's a club.
00:24:08.880 --> 00:24:18.640
When I first entered the industry, um similar time period to yourself, you you was you did feel like you were kind of treading on people's toes coming into coming into this world.
00:24:18.799 --> 00:24:24.559
Uh I do think it's much more collaborative now, but there's still so much, so much work that we can do.
00:24:25.200 --> 00:24:30.880
And final question: if you could give one piece of advice to anyone looking to work in the sector, what would it be?
00:24:31.359 --> 00:24:37.119
To work in the sector, I think uh just just to be inquisitive.
00:24:37.279 --> 00:24:42.720
Um there's there's so much to learn um it all different aspects.
00:24:43.200 --> 00:24:46.319
So I guess don't be afraid to not know something.
00:24:46.480 --> 00:24:56.720
Um walk in with the questions, and then and then from that you'll figure out where you add the most value, and then just go towards that because we we need great people, we always need great people.
00:24:56.799 --> 00:25:02.720
Um, and those are the ones that come with with a willingness to learn and and and help and lean in.
00:25:03.039 --> 00:25:08.400
Yeah, I'll say I think the the most uh give and answer is is inquisitive curiosity.
00:25:08.559 --> 00:25:13.119
I this industry is the perfect place for people with that curious mind.
00:25:13.279 --> 00:25:16.799
You can go far in this game if if you ask the right questions.
00:25:17.119 --> 00:25:18.240
Or take up cycling.
00:25:18.400 --> 00:25:22.000
That's the take up cycling and come join me and can every year.
00:25:22.240 --> 00:25:24.079
That's the other bit of advice I'd give.
00:25:24.400 --> 00:25:30.240
I was gonna say my advice is join Data Centre Cycling Club, uh, link in the comments, go find it.
00:25:30.400 --> 00:25:31.279
Join Tom and Cann.
00:25:31.440 --> 00:25:32.160
Don't join me and can.
00:25:32.400 --> 00:25:36.079
I'll I'll um I'll I'll wave you in when you arrive and buy you a drink.
00:25:36.160 --> 00:25:40.880
But you certainly won't catch me on a on my BMX heading uh heading around Cairn.
00:25:41.119 --> 00:25:42.559
Um thanks for your time.
00:25:42.640 --> 00:25:44.160
Good luck with Adapt for MDC.
00:25:44.480 --> 00:25:47.119
Again, like you know, congratulations on the on the launch.
00:25:47.200 --> 00:25:49.759
Obviously, looking forward to see your success over the coming years.
00:25:49.920 --> 00:25:56.400
We'll put the link in the show notes if anyone wants to learn anything more about the company, about Tom and about the team, that'll be there.
00:25:56.480 --> 00:26:03.839
And we'll also put the link in for the cycling club so that if anyone is a keen rider out there, they can uh come and talk to you on a bike somewhere.
00:26:04.160 --> 00:26:04.640
Awesome.
00:26:04.720 --> 00:26:06.480
Thanks, and I appreciate your time, mate.
00:26:06.880 --> 00:26:08.079
Pleasure, good to catch up.
00:26:08.160 --> 00:26:10.799
That's it for this episode of the Inside Data Centre Podcast.
00:26:10.880 --> 00:26:11.839
Thanks for tuning in.
00:26:12.000 --> 00:26:19.920
This show is brought to you by DataX and Data Exec, the recruitment partners powering global talent across the data centre and digital infrastructure world.
00:26:20.079 --> 00:26:24.000
If you enjoyed today's conversations, don't forget to follow, like, and subscribe.
00:26:24.160 --> 00:26:26.559
It really helps us grow the podcast community.
00:26:26.880 --> 00:26:32.160
Stay tuned for more stories, insights, and conversations with the people shaping the future of our industry.
00:00:00.160 --> 00:00:06.080
The data centre industry is moving faster than ever, and behind the scenes some seriously smart people are making it all happen.
00:00:06.240 --> 00:00:13.039
Welcome to the Inside Data Centre Podcast, where we dive into the stories, insights and innovations driving real change across the industry.
00:00:13.279 --> 00:00:19.920
From hyperscale campuses to game-changing technologies, we're talking to the people shaping the future of digital infrastructure.
00:00:20.079 --> 00:00:25.839
Whether you live in breathed data centres or just want to understand what's powering our connected world, you're in the right place.
00:00:26.079 --> 00:00:27.920
Welcome to the Inside Data Centre Podcast.
00:00:28.000 --> 00:00:33.840
On this episode, I'm joined by Tom Babbington, founder and CEO of Adaptive MDC.
00:00:34.159 --> 00:00:35.200
Good afternoon, Tom.
00:00:35.439 --> 00:00:37.039
Hey Ali, how are you going, mate?
00:00:37.359 --> 00:00:37.840
I'm good.
00:00:38.000 --> 00:00:39.039
Second time on the podcast.
00:00:39.119 --> 00:00:44.000
I had to look back before we uh started this conversation to see when it was, two years ago, 2024.
00:00:44.240 --> 00:00:46.640
But that time we were talking about a very different subject.
00:00:46.799 --> 00:00:49.679
We were talking about Data Centre Cycling Club.
00:00:49.840 --> 00:00:54.240
So I think uh we're we're in a slightly different area remit now for this conversation.
00:00:54.560 --> 00:01:00.159
Yeah, I I happily uh talk about cycling, but it it it it might not get the same uh same traction.
00:01:00.399 --> 00:01:01.439
Yeah, it was.
00:01:01.920 --> 00:01:02.560
Yeah.
00:01:03.039 --> 00:01:05.840
And I think um data center cycling club's still going as well, obviously.
00:01:05.920 --> 00:01:06.719
We're still good?
00:01:06.879 --> 00:01:09.120
Yeah, so we did we did the ride in CAN this year.
00:01:09.280 --> 00:01:13.120
Um we were less ambitious, more inclusive, which has been the feedback.
00:01:13.200 --> 00:01:20.319
So we did I think the the the we did four days and we did 150, 120, 150k average each day.
00:01:20.400 --> 00:01:23.280
So there's 20 riders, so it was bigger.
00:01:23.519 --> 00:01:29.920
Um we raised um plenty of money, which which goes to Action Medical.
00:01:30.000 --> 00:01:31.920
Um yeah, it was it was a good crack.
00:01:32.239 --> 00:01:32.879
Yeah, well done.
00:01:32.959 --> 00:01:36.319
And obviously, if anyone wants to get involved in the future, I'm sure they'll be able to find it.
00:01:36.400 --> 00:01:39.760
Data center cycling club has a website and everything, so it should be easy to find.
00:01:39.840 --> 00:01:41.040
But we'll put it in the show notes as well.
00:01:41.120 --> 00:01:41.680
You know, you never know.
00:01:44.480 --> 00:01:46.560
There's always some, always some lurking.
00:01:46.719 --> 00:01:48.480
You can't you can't hide for them.
00:01:48.799 --> 00:01:49.519
Definitely.
00:01:49.680 --> 00:01:56.079
And back to the the world of um work and and data centers, not that cycling is not work, it's hard work, not that I cycle.
00:01:56.159 --> 00:01:59.760
But um, you've recently launched Adaptive MDC.
00:01:59.840 --> 00:02:05.439
So, say we're in a very different position to where we were a couple of years ago when we had a conversation from a from a work perspective.
00:02:05.599 --> 00:02:10.080
Do you want to give a bit of an overview and just explain how you've arrived at this point?
00:02:10.400 --> 00:02:11.919
Yeah, um absolutely, thank you.
00:02:12.000 --> 00:02:22.159
Yeah, so so adaptive uh MDC is uh a modulat business where we take the traditional data center problem and turn them into modular solutions.
00:02:22.240 --> 00:02:26.879
So everything from design, manufacturing, construction, integration through to commissioning.
00:02:26.960 --> 00:02:33.360
Um, so like the full life cycle um where we try and drive through efficiency throughout the whole process.
00:02:33.520 --> 00:02:43.520
Um lots of things, so uh you know, AI has changed, the density sort of uh the the these facilities having smaller footprints lends itself to modular.
00:02:43.599 --> 00:02:54.400
Um, and we're we're looking sort of just just improving sort of the execution of the data center delivery um or optimizing it as best we can.
00:02:55.120 --> 00:02:59.039
And your background in the industry, how did it kind of lead you to this point?
00:02:59.120 --> 00:03:02.400
Because I guess I think a lot of the time we learn from experience, don't we?
00:03:02.560 --> 00:03:08.159
We see these challenges, and then some people create a solution which becomes an organization, which is obviously what you've done.
00:03:08.240 --> 00:03:13.039
But what how is your experience sort of in the world of data centers led you to this point?
00:03:13.680 --> 00:03:23.039
Uh so I I've been thinking about this a lot as sort of my route to the data center market is probably quite traditional in that I never planned on it happening.
00:03:23.280 --> 00:03:25.199
Um, I just sort of evolved into it.
00:03:25.360 --> 00:03:33.759
My construction career started in Afghanistan um in 2006, and we were building uh facilities for the vulnerable in the community.
00:03:33.919 --> 00:03:38.400
Um then I came back and had my one and only construction project in the UK.
00:03:38.560 --> 00:03:48.719
So six years then six years after that, I I'm on the hyperscale uh job in Finland um looking after uh uh elements of delivery for that.
00:03:48.800 --> 00:03:56.639
So I've I've spent the majority of my career working overseas, I think about 42 countries and counting in in different locations that I've worked.
00:03:56.960 --> 00:04:11.360
Um I am happiest when I'm in a room of of people from different cultures trying to understand what their problems are, how to solve them, um, rather than just sort of sticking to what the book says, the traditions of this is how it's done.
00:04:11.439 --> 00:04:13.520
It's like the the world isn't made up of that.
00:04:13.759 --> 00:04:21.839
It's made up of of problems that that need to be solved, and and and ultimately people people create them, so you've got to understand what it is that they they need solving.
00:04:22.319 --> 00:04:31.600
Yeah, and I think from the culture perspective, it's quite an interesting point as well, because this industry is full of different cultures, diverse, you know, like different people on these projects.
00:04:31.839 --> 00:04:37.360
Is that you'll say you've got probably more experience than most working and operating in that environment?
00:04:37.600 --> 00:04:46.639
But do you think that it's important for people to be successful in this industry if they have that ability to kind of flex and mould to all these different cultures?
00:04:47.040 --> 00:04:52.639
Well, I think uh inevitably the greatest thing about our industry is the people that are in it, right?
00:04:52.720 --> 00:04:56.560
You know, that's that's how it's been able to grow and be so successful around the innovation.
00:04:56.720 --> 00:05:00.800
But where we're struggling now is finding enough good people to execute.
00:05:01.120 --> 00:05:06.319
So, you know, for years we've talked about a resourcing problem, there's a capacity problem.
00:05:06.480 --> 00:05:14.560
So, yeah, you've got to you've got to sort of see what opportunities lay ahead in terms of the people and the capacity and what they can do and find out how they can solve it.
00:05:14.720 --> 00:05:25.920
So, yeah, um, you know, there's there's been times where I've been in we had a I was working on a hotel uh construction project in Tunisia and we couldn't get the guys to uh work any later.
00:05:26.079 --> 00:05:33.199
So we were like paying them in advance and getting them to work later because the the job was running late and they would take the money and not turn up for another week.
00:05:33.360 --> 00:05:35.759
So we decided not to pay them, then they didn't turn up.
00:05:35.920 --> 00:05:38.319
So we just said, Well, what are they missing?
00:05:38.399 --> 00:05:39.759
We we cook them dinner.
00:05:39.839 --> 00:05:40.879
So it's simple as that.
00:05:41.040 --> 00:05:42.959
Every night there was dinner for all the staff.
00:05:43.040 --> 00:05:54.560
They ended up bringing all their families, but they stayed until the the day was done, and it it improved productivity, and and it's a it's a it's a small part of of how we how we do it.
00:05:54.720 --> 00:06:07.120
Um so I think this industry has to has to find a different way rather than just going out and recruiting straight from university or straight from primary school now, because it's like you've got to get the good people in, you've got to find a different way.
00:06:07.680 --> 00:06:09.279
Yeah, 100% interesting.
00:06:09.439 --> 00:06:18.959
And on the modular point, which is obviously why we're here, I think it feels like I've been talking about modular data centers for the whole time of since I started this podcast, which is coming up to six years.
00:06:19.199 --> 00:06:32.879
But I think now the the conversations are becoming more kind of detailed and and people it's a real focus now for companies, and it's a solution that the industry has to you utilize in order to kind of deliver at the scale that that it is.
00:06:33.040 --> 00:06:41.839
But you've do you think sort of from your perspective that now is the time where the industry's really pushing towards this new sort of this model of working?
00:06:42.319 --> 00:06:45.680
Uh yes, um, of course.
00:06:45.759 --> 00:06:51.199
Uh, but I don't think fundamentally anything's really changed in the industry.
00:06:51.360 --> 00:06:58.079
I think the demands are still high, it's still great, like there's still an expectation to deliver something in a in as quickly as possible.
00:06:58.319 --> 00:07:04.480
I think it's the biggest change is the amount of people asking for it to be delivered in this unachievable time.
00:07:04.639 --> 00:07:13.759
So customers have always said we need this online in 12 months, but now it's the number of customers is exponential, so we have to find alternatives.
00:07:13.839 --> 00:07:23.920
So and you know, innovation has always happened, and we're an industry full of great engineers who are coming up with solutions to solve these problems.
00:07:24.240 --> 00:07:29.759
It's finding the um the way of putting them into practice.
00:07:30.000 --> 00:07:31.680
I keep saying it's like the execution.
00:07:31.839 --> 00:07:41.600
How can we how can we standardize a certain design to be able to solve to to to deliver that?
00:07:41.759 --> 00:07:44.560
Um and and that's where modular lends itself.
00:07:44.800 --> 00:07:56.240
Plenty of companies are doing different elements of modular, whether it's their their sort of the productizing, their components, and it's bringing them together to make sure you can just control as much as possible.
00:07:56.480 --> 00:08:02.480
Now I'm not saying it's standard, you have to design a solution to meet the customer's requirements.
00:08:02.560 --> 00:08:03.439
That's not going to change.
00:08:03.680 --> 00:08:21.120
They will never buy a data center just off the shelf because the nuance of what they want to go in it or how they want it to operate will change, and and that's part of what what our approach is is making sure the design is is sort of scope-specific, but optimising what's already available.
00:08:21.759 --> 00:08:32.639
You touched on innovation as well, and I I guess what we've seen a lot in the industry over the last um say two years is a lot of new innovative organizations enter the industry.
00:08:32.799 --> 00:08:44.159
And I think what that's brought is a different perspective and looking for different ways of working rather than the legacy data center market that kind of had a way of doing things and that was just how how we did things.
00:08:44.320 --> 00:08:50.000
Do you think that is also impacting on execution or how we build and how we deliver data centers?
00:08:50.720 --> 00:08:54.639
Well, yeah, I think I think they both they both there is a place for both, right?
00:08:54.720 --> 00:08:56.399
So it's it's quite a broad market.
00:08:56.559 --> 00:09:06.159
There's you know the the location of the data centers will dictate the sort of how they're gonna build it, but the speed and scale is is the alternative.
00:09:06.320 --> 00:09:17.360
So um some of the biggest jobs, construction jobs happening at the moment are going to be modular in in the Middle East, and that's just because they can control so much so many elements of it.
00:09:17.519 --> 00:09:26.000
But the smaller market where concrete is king and they've got workforces in place, of course, they can go and and use the traditional method.
00:09:26.320 --> 00:09:34.480
So it doesn't solve every problem, it just provides an alternative that allows us to be to be much more efficient with the resources that are available.
00:09:34.639 --> 00:09:56.480
And there is there is plenty of manufacturing capabilities across the across the world that can help provide modular solutions, is just having the expertise around the the data center side of it to make sure the quality and the standardizations within it are suitable for our very demanding but uh but lovely customers.
00:09:56.960 --> 00:09:58.080
Very lovely customers.
00:09:58.240 --> 00:10:00.720
And has geo geography impacted as well?
00:10:00.879 --> 00:10:17.279
Because again, I if I if I think back over the conversations I've had over the years, a lot of the conversation when we were talking about modular or similar type solutions was the ability to build in more remote regions where there, like you said, there isn't a workforce or there isn't the access that they need.
00:10:17.360 --> 00:10:21.919
Is that again, is that impacted the the sort of the move further towards modular?
00:10:22.240 --> 00:10:23.600
Yeah, absolutely, right.
00:10:23.759 --> 00:10:38.799
Uh it it you know, you you want to get a thousand people to work up in the north of Finland, uh, or you build it in a in a controlled environment somewhere else in Europe and ship it to play so you're only on the ground for for a short period of time.
00:10:38.960 --> 00:10:46.000
You can get testing done, or majority of the testing done in a facility off site and then shift it to site so there's less time on the ground.
00:10:46.159 --> 00:10:52.720
It controls health and safety aspects, it controls um the movement of materials in and out of of the same area.
00:10:52.960 --> 00:11:03.039
So it's not it's not the same as um you know we've seen in the past of pick a random location where land and power is cheap and try and get a thousand people over there for 18 months.
00:11:03.519 --> 00:11:06.799
That that's that's gonna have to change.
00:11:07.440 --> 00:11:11.600
Yeah, they are there are still some trying it though, I will I will say that.
00:11:11.840 --> 00:11:18.720
Um from a and and customer, so when you're in your conversations with customers, again, is this is is it now this is a solution they're looking for?
00:11:18.879 --> 00:11:22.799
This is not a solution that they may you know proceed with at some point.
00:11:22.879 --> 00:11:28.639
This is something they need now in order to achieve their sort of unbelievable delivery schedules.
00:11:28.879 --> 00:11:33.759
Yeah, look, it's it's where we're spending a lot of our time is developing reference designs for customers.
00:11:33.919 --> 00:11:48.320
So they have a standard go-to-market solution where depending on the location and depending on slight variances in terms of what what what deployment they want, we can we can change our design.
00:11:48.480 --> 00:11:58.159
So reference designs from everything from five kilowatt racks up to 130 at the moment, but can go up all the way to 300, 600 when that comes about.
00:11:58.320 --> 00:12:16.480
It's about configuring the design to be fully modular, so it's not just a kit of parts, it's it's looking at it from a construction commissioning standpoint and how you bring everything in together so your your shipping is as efficient as possible, um, and not just not just boxes of air and putting everything in.
00:12:17.840 --> 00:12:21.360
What scale can we are we talking from a modular perspective now?
00:12:21.440 --> 00:12:36.639
Because again, I I I think the perception to the outside world, or I might it might just be people like myself, the uneducated, that you know, modular they assume smaller, you know, like our smaller facilities, you know, let's say we can do a one megawatt haul and a modular perspective build.
00:12:36.720 --> 00:12:41.600
What scale are we talking now when you're when you're working with these customers for their designs?
00:12:41.840 --> 00:12:46.799
So uh uh limitless uh is it because it is.
00:12:46.960 --> 00:12:49.120
Um how big do you want to go?
00:12:49.200 --> 00:12:52.480
It it we of course there's the one megawatt.
00:12:52.639 --> 00:12:56.480
There's the containerized solution, which is a data center in a container.
00:12:56.639 --> 00:13:05.120
Um, our solution is a modular approach where it's a series of containers that make up a sort of a modular um solution.
00:13:05.279 --> 00:13:12.000
So there's plenty of jobs over 200 megawatts that are modular at the moment that that I know of.
00:13:12.240 --> 00:13:17.919
Um some of the sizes that we're working on are over 100 megawatts.
00:13:18.000 --> 00:13:21.440
Um, but then there is a repeatable two and a half, a repeatable one.
00:13:21.759 --> 00:13:26.000
Um there's a deployment in a in a powered shell of different configurations.
00:13:26.159 --> 00:13:29.039
So it is truly it's limitless.
00:13:29.600 --> 00:13:41.279
Yeah, and I and I think that again is it's sort of busting that perception that I've I always think people think modular is small, like it's strange, it's just the word, I think the terminology leads people to that to that route.
00:13:41.360 --> 00:13:49.519
So again, as as the industry scales, it needs these solutions in order to scale because it can't keep up with the demand and deliver quick enough.
00:13:49.840 --> 00:14:00.000
And we we so we always open our our introductory presentation with uh this is what module data center means to us, and that is it's not containerized.
00:14:00.960 --> 00:14:06.320
There was an example of uh a containerized data center in the footsteps of of the conference at Cannes year.
00:14:06.480 --> 00:14:26.080
That for us is containerized, not modularised, that's just a container that that sir solves a problem, but scaling that to a level where you're trying to get uh you know a huge sort of a gigawatt deployment or 250 megawatt deployment, you have to do a design that is more efficient because that would not be.
00:14:26.960 --> 00:14:27.840
Yeah, definitely.
00:14:28.000 --> 00:14:36.559
And I think if and if you talk about CAN, I think a lot of the outside of the uh the drinks and everything, a lot of the the conversations were around execution.
00:14:36.639 --> 00:14:41.519
And I I genuinely think execution's now become one of the key challenges in this market.
00:14:41.679 --> 00:14:56.879
I think execution and community, you know, we've had power and people for years, but I really think now the conversations I'm having anyway, with with whether it's with funds, investors, CAs, whatever they might be, is around execution and also the community and managing that challenge.
00:14:56.960 --> 00:15:03.200
But do you think the industry is doing enough to really sort of solve this execution problem?
00:15:04.240 --> 00:15:09.200
Um uh no, probably not, uh if I'm honest.
00:15:09.360 --> 00:15:24.639
I think uh talking about it a lot, um, but we all get very excited about the next shiny object, the next big project that's gonna take up all the the power in the land and how we're gonna deliver it um is is sort of secondary to it.
00:15:24.799 --> 00:15:27.120
So I think I think we can certainly do much more.
00:15:27.200 --> 00:15:37.600
I think there can be plenty of collaboration within the industry around how we how we combine forces to to to make sure that execution is paramount.
00:15:37.840 --> 00:15:43.039
Um everyone if everyone stays in their lane, then then nothing changes, right?
00:15:43.120 --> 00:15:51.200
That the execution doesn't change, everyone's just trying to sell their their product because that's obviously how their businesses are stacked up.
00:15:51.360 --> 00:16:06.559
But but you combine that and you bring that all together and you're sort of agnostic in your delivery, or you um have a design that integrates different components from different people because they have capacity online or they have sorry capacity within their their production lines when it's needed.
00:16:06.799 --> 00:16:14.879
So rather than sticking to one that's saying it's 52 weeks, you can't have a generator of 52 weeks, or your switch gear is you know it's it's 18 months.
00:16:14.960 --> 00:16:24.720
It's like, okay, well, if we can now get a customer the same specification in nine months, why would we not do that to get their GPU online?
00:16:24.879 --> 00:16:30.720
Like it still it still serves the same same uh or solves the same problem.
00:16:30.879 --> 00:16:36.720
And I think we we have to um we have to sort of look beyond the next three years.
00:16:36.879 --> 00:16:38.799
Capacity's tied up for the next three years, you know.
00:16:38.879 --> 00:16:40.159
Look at production lines.
00:16:40.320 --> 00:16:48.000
You you you you can't now go to uh many vendors and say, what's your timeline without them saying it's a year.
00:16:48.240 --> 00:16:52.480
How do you how do you get certainty around a any timeline with a vendor?
00:16:52.559 --> 00:16:54.399
It's you go to them with an order.
00:16:54.639 --> 00:16:55.759
That's the only way.
00:16:55.919 --> 00:17:04.160
Um or you've got a deep enough pocket that you can just pay over the odds, which which isn't healthy, which isn't which isn't healthy for the market.
00:17:04.319 --> 00:17:14.400
So I think if you if you go with that collaborative mindset of of trying to solve the industry problem, you'll solve your customer's problem.
00:17:14.960 --> 00:17:15.680
Yeah, and I agree.
00:17:15.839 --> 00:17:19.920
And you have seen, I guess, more flexibility in customer supply chain, haven't you?
00:17:20.000 --> 00:17:33.359
You've seen them have to increase their supply chain in certain areas, not not broadly across everything, but in certain areas they've had to increase their supply chain in order to achieve part of their their targets.
00:17:33.519 --> 00:17:39.119
They're still not there where they need to be, but there is a bit more flexibility now than there was going back two or three years.
00:17:39.359 --> 00:17:40.240
Yeah, I think so.
00:17:40.799 --> 00:17:42.559
There's less rigidity for sure.
00:17:42.640 --> 00:17:50.160
Uh look, there's plenty more alternatives, and you know, the the quality's in general, the quality is improving.
00:17:50.400 --> 00:17:57.359
Um so yeah, I I think I think that's been a real positive is is just the breadth of what's available.
00:17:57.599 --> 00:18:06.319
Um but still it still needs to be delivered uh and and and uh sort of the the execution still needs to be focused on.
00:18:06.960 --> 00:18:07.680
Yeah, definitely.
00:18:07.920 --> 00:18:15.039
And when you look to the future, which is it always seems impossible in this industry because the future seems to arrive pretty much every day.
00:18:15.200 --> 00:18:21.279
Um what what does a data centre look like in you know, one, two, five, pick a pick a number of years?
00:18:21.440 --> 00:18:24.880
Um I think again, like we've seen that evolution, haven't we?
00:18:25.039 --> 00:18:32.799
And data centres are very different uh buildings or facilities to what they were, but you how do you see it progressing over the next few years?
00:18:33.680 --> 00:18:39.279
Yeah, uh so I I thought all data centers are going to go to space, isn't that what Space SpaceX is hoping for?
00:18:39.440 --> 00:18:47.920
Um I think I I read I read most weeks that they are going to space, and then two articles later on my search, it says they're not going to space.
00:18:48.000 --> 00:18:50.240
So I think we're still a bit unknown on that one.
00:18:50.400 --> 00:18:51.039
Depends who you are.
00:18:51.119 --> 00:18:54.960
So what depends if you own if you own a space company, then you say they are.
00:18:55.599 --> 00:19:00.480
Yeah, so so once you've got your telescope out, you can see what the space, the data sensor of the future look like.
00:19:00.640 --> 00:19:04.799
No, look, I I uh it's it's it's an impossible question.
00:19:04.960 --> 00:19:15.119
I think you know about 100 kilowatt racks happening now, 150 are happening now, 600 are just around the corner, um direct to chip cooling is gonna happen.
00:19:15.279 --> 00:19:17.279
All the technologies there, the innovations happening.
00:19:17.359 --> 00:19:25.039
I think what we'll see differently in the future is um the public perception, or I hope.
00:19:25.119 --> 00:19:28.640
I but uh you know the public perception of data centers is gonna have to change, right?
00:19:28.799 --> 00:19:33.680
They're gonna have data centers are gonna have to become part of economic ecosystems.
00:19:34.000 --> 00:19:34.799
It just has to happen.
00:19:34.880 --> 00:19:43.359
It's not they're not a drain on power, they're not a drain on water, not a drain on all the the resources, taking it away from municipalities and and and the people.
00:19:43.680 --> 00:19:45.279
They will will be part of those.
00:19:45.440 --> 00:19:53.359
It allows people to work from home, allows them to work more efficiently, it it gives jobs, it adds uh value to to those environments.
00:19:53.599 --> 00:19:55.039
And I think that'll be the change.
00:19:55.119 --> 00:20:03.039
So what they'll look like from the outside, I don't know, but hopefully people's perception of what they look like and what they are will shift massively.
00:20:03.599 --> 00:20:05.359
Yeah, and I think it's a really important point.
00:20:05.440 --> 00:20:13.759
And and I guess in your career, you you've been in the industry for a let's say for a while compared to a lot of people, even though it's not that long.
00:20:14.240 --> 00:20:28.880
The the the perception of the industry, kind of what I feel is that we've gone from no one really knowing, no one really caring, to all of a sudden it feels like everybody really caring for different reasons.
00:20:28.960 --> 00:20:34.880
It's not related to a lot of it's related to energy or land or what and I completely understand that that viewpoint as well.
00:20:34.960 --> 00:20:36.480
But have you do you feel the same?
00:20:36.640 --> 00:20:41.200
Do you feel like there's just been this sudden shift from not in the news to front page news?
00:20:41.519 --> 00:20:56.319
Yeah, uh I just so yesterday I was talking to to uh a friend in our office and just talking through sort of a an investor deck around uh a funds development for day centers.
00:20:56.400 --> 00:21:03.839
And I was talking through just just basic rule of thumb metrics for day center construction, and he was like, How expensive?
00:21:04.000 --> 00:21:04.480
How much?
00:21:04.640 --> 00:21:14.400
I'm like, yeah, but and I explained why it was that much, explain why the and it's become it's come every day people asking questions about day centers.
00:21:14.640 --> 00:21:23.599
And that was you know, we we met people um actually it was in Cannes, they came over, they have a they have a power station and they're like, I've got power, I've got land, how do I build a day center?
00:21:23.680 --> 00:21:24.799
And I'm like, great.
00:21:24.960 --> 00:21:27.279
You know, it's become ever present in everyone's mind.
00:21:27.359 --> 00:21:44.079
It sort of it's not I I I I think that's that's the biggest shift is it's no longer just explaining that they are the power stations of the future in in that sense, it's it's driving economic change that that has been ignored so far.
00:21:44.240 --> 00:21:46.480
Um so yeah.
00:21:53.440 --> 00:21:59.759
And you end up doing their job by finding the off taker at the end because you you know, fifteen years of experience.
00:22:00.079 --> 00:22:02.319
In the sector, you get to know a lot of people.
00:22:02.640 --> 00:22:06.319
Um, so I mean, I'm guessing you're pretty much the same space.
00:22:06.400 --> 00:22:11.279
It's like how many people come to you and say, 'I've got some power and some land, Andy.' Who do I need to talk to?
00:22:11.759 --> 00:22:12.559
Exactly that.
00:22:12.640 --> 00:22:16.079
Yeah, it's um farmers very often in the UK.
00:22:16.400 --> 00:22:17.839
I speak to a lot of farmers.
00:22:17.920 --> 00:22:19.279
I had a lot of messages yesterday.
00:22:19.359 --> 00:22:30.559
I posted about the new the new data centre in Devon, and I had a lot of messages from residents of Devon asking me why I'm building my data centre sat down there, and I had to explain it's not my data centre.
00:22:31.200 --> 00:22:32.799
I'll just I'll just sharing the news.
00:22:32.880 --> 00:22:33.759
That's that's all I had.
00:22:33.839 --> 00:22:35.200
But yeah, no, it's fascinating, though.
00:22:35.279 --> 00:22:40.640
I do agree, and I think it's all nice to educate, it's it's totally our our challenge to solve solve the perception piece.
00:22:40.720 --> 00:22:49.839
But it's um it's a very, very interesting journey to go from, as I say, like I felt I had to kind of tell everybody about what I did and really explain it.
00:22:50.079 --> 00:22:54.480
You get to a point where you're like, oh, I can't, yeah, I just can't be bothered to go into this detail anymore.
00:22:54.640 --> 00:22:55.599
You don't really care, do you?
00:22:55.680 --> 00:22:58.160
To now everyone's saying, you work in data centres, don't you?
00:22:58.240 --> 00:22:59.200
Like, what what do you do?
00:22:59.279 --> 00:22:59.839
What what are they?
00:22:59.920 --> 00:23:01.200
They're gonna build one near our house.
00:23:01.279 --> 00:23:04.720
You know, it's it's fascinating, but a lot of work to do for the industry.
00:23:04.880 --> 00:23:10.079
Um before we close up, just a couple of quick fire questions, which I ask everyone on the pod.
00:23:10.480 --> 00:23:15.599
If you could ask everyone in the sector to start or stop doing one thing, what would it be?
00:23:16.480 --> 00:23:18.160
Oh, stop doing something.
00:23:18.400 --> 00:23:22.720
Uh trying to think of what people do that that bothers me.
00:23:22.960 --> 00:23:30.079
No, I think I think uh uh I think just stop accepting that that's how it's always been done, right?
00:23:30.160 --> 00:23:33.200
We're trying to move at a speed that we have to find alternatives.
00:23:33.359 --> 00:23:40.240
So uh you know that goes to what I was saying about collaborating, you know, across disciplines and companies and borders.
00:23:40.400 --> 00:23:47.839
I think there's much more we can do if you just if you just sort of stop accepting sort of the normal way of doing things.
00:23:48.079 --> 00:23:51.680
Uh is that a good enough answer?
00:23:52.079 --> 00:23:52.880
That is a good answer.
00:23:53.200 --> 00:23:54.000
I'd be more brutal.
00:23:55.359 --> 00:23:55.839
Yeah, exactly.
00:23:55.920 --> 00:23:56.400
It's different.
00:23:56.480 --> 00:23:57.119
Where's the line?
00:23:57.200 --> 00:24:02.640
But it's uh no, it's a fascinating, it's an interesting point as well, because I feel like the industry is going through that shift now.
00:24:02.720 --> 00:24:08.720
I I do I remember when I first joined, well not joined, but first started where it's not as I make it out like it's a club.
00:24:08.880 --> 00:24:18.640
When I first entered the industry, um similar time period to yourself, you you was you did feel like you were kind of treading on people's toes coming into coming into this world.
00:24:18.799 --> 00:24:24.559
Uh I do think it's much more collaborative now, but there's still so much, so much work that we can do.
00:24:25.200 --> 00:24:30.880
And final question: if you could give one piece of advice to anyone looking to work in the sector, what would it be?
00:24:31.359 --> 00:24:37.119
To work in the sector, I think uh just just to be inquisitive.
00:24:37.279 --> 00:24:42.720
Um there's there's so much to learn um it all different aspects.
00:24:43.200 --> 00:24:46.319
So I guess don't be afraid to not know something.
00:24:46.480 --> 00:24:56.720
Um walk in with the questions, and then and then from that you'll figure out where you add the most value, and then just go towards that because we we need great people, we always need great people.
00:24:56.799 --> 00:25:02.720
Um, and those are the ones that come with with a willingness to learn and and and help and lean in.
00:25:03.039 --> 00:25:08.400
Yeah, I'll say I think the the most uh give and answer is is inquisitive curiosity.
00:25:08.559 --> 00:25:13.119
I this industry is the perfect place for people with that curious mind.
00:25:13.279 --> 00:25:16.799
You can go far in this game if if you ask the right questions.
00:25:17.119 --> 00:25:18.240
Or take up cycling.
00:25:18.400 --> 00:25:22.000
That's the take up cycling and come join me and can every year.
00:25:22.240 --> 00:25:24.079
That's the other bit of advice I'd give.
00:25:24.400 --> 00:25:30.240
I was gonna say my advice is join Data Centre Cycling Club, uh, link in the comments, go find it.
00:25:30.400 --> 00:25:31.279
Join Tom and Cann.
00:25:31.440 --> 00:25:32.160
Don't join me and can.
00:25:32.400 --> 00:25:36.079
I'll I'll um I'll I'll wave you in when you arrive and buy you a drink.
00:25:36.160 --> 00:25:40.880
But you certainly won't catch me on a on my BMX heading uh heading around Cairn.
00:25:41.119 --> 00:25:42.559
Um thanks for your time.
00:25:42.640 --> 00:25:44.160
Good luck with Adapt for MDC.
00:25:44.480 --> 00:25:47.119
Again, like you know, congratulations on the on the launch.
00:25:47.200 --> 00:25:49.759
Obviously, looking forward to see your success over the coming years.
00:25:49.920 --> 00:25:56.400
We'll put the link in the show notes if anyone wants to learn anything more about the company, about Tom and about the team, that'll be there.
00:25:56.480 --> 00:26:03.839
And we'll also put the link in for the cycling club so that if anyone is a keen rider out there, they can uh come and talk to you on a bike somewhere.
00:26:04.160 --> 00:26:04.640
Awesome.
00:26:04.720 --> 00:26:06.480
Thanks, and I appreciate your time, mate.
00:26:06.880 --> 00:26:08.079
Pleasure, good to catch up.
00:26:08.160 --> 00:26:10.799
That's it for this episode of the Inside Data Centre Podcast.
00:26:10.880 --> 00:26:11.839
Thanks for tuning in.
00:26:12.000 --> 00:26:19.920
This show is brought to you by DataX and Data Exec, the recruitment partners powering global talent across the data centre and digital infrastructure world.
00:26:20.079 --> 00:26:24.000
If you enjoyed today's conversations, don't forget to follow, like, and subscribe.
00:26:24.160 --> 00:26:26.559
It really helps us grow the podcast community.
00:26:26.880 --> 00:26:32.160
Stay tuned for more stories, insights, and conversations with the people shaping the future of our industry.
