 Welcome to JSA TV and JSA Podcast, the newsroom for telecom and data center professionals. I'm Jean-Marc Slim and joining me today is Marc Seymour, CTO of Future Facilities. And Marc, it's a pleasure to speak to you after such a long time. Before we jump into what you do and the notion of digital twins, which is quite interesting, especially in the data center space. Tell us how did you get involved with the data center marketplace in the first instance? Okay, well, I'm a simulation professional, I guess. I studied numerical mathematics and soon found that simulation was a great way of understanding physical behaviors that we otherwise had no idea about. And I was using that initially in the defense industry and then I moved out into using it for the built environment when people started realizing that we could do things with simulation that would otherwise always have been done by experiments, part scale models, full scale models of buildings for prestigious buildings and things like that. And it became obvious that we needed tools that were specialized to these particular applications. And I was using general purpose software in the building industry that was commercially available, but it was really difficult to use. And we realized that to model buildings, we needed something specific. And in talking to the people that were providing that software, we came to the conclusion we need these specialist tools and it turned out that they were setting up a new company to focus on specific tools. So I got involved with specifying their new tools for the building industry, but they also had a major line for electronics cooling. And that electronics cooling was an important connection because we started realizing as we developed the tools that the electronics problem was breaking out into the data center. So what I mean by the electronics problem is that basically, if you don't manage the thermal performance, you don't have an electronics box, your phone catches fire or whatever. Small detail. Yeah. So, I mean, obviously that's an extreme example, but the same, it's kind of like a fire for a data center, because if your equipment overheats, even if it doesn't shut down, which it may do, it will probably slow down. And if your transactions disappear, then you haven't got business and data centers are the heart of the business. And so it became obvious that this heat problem was bursting out into the data center. And so we realized that we should apply the technologies of predicting the future when you have no data center to the data center. And importantly, the difference between what we were seeing compared with the traditional use of simulation was that data centers were unusual in that they offered a lifecycle opportunity. Most predictive simulation is used for design cycle and design cycle alone. But for the data center, the one thing you know about it is that what you think it's going to be like in design is something you'll never see in practice. Because the electronics box you designed for 20 years and the components that go in at the computers are changing all the while, and over that lifetime, there'll have to be many cycles. So what you actually implement is completely different. So data centers seemed obvious and an important subject for this thing that I get a great kick out of which is being able to predict that, you know, something that's complex and understand what's going to happen when you can, you know, it's pretty hard to imagine how that complex process works in detail, but but computers can do that. I mean, it's extraordinary. And I love the full cycle journey of where you started and how you've got involved with data centers. And I mean, what you do is still very unique. I don't actually remember when you start to future facilities, but it's been seven, eight years. And it's pretty much longer than that. We formed future facilities in 2004. 2004. Wow. Okay. Yeah, I was, I didn't even know what it was by then. So we were our problem was that our vision was good. We're still on the same vision today that we had back in 2004. But we were just way ahead of the market. The market understood the concept of using simulation for design, operation. No, no clue at all. And it's it took us a long, long while, which is why you've got the perception of seven or eight years. It took us a long, long while to get from just offering devices and simulation software for point solution in design to simulation for operation, which is now, you know, a really big, big thing because, you know, the size of these data centers and the difficulty of managing them is, is severe. And you need tools to be able to understand what will actually happen because you're making changes on big scales and you can't undo them. So once you've set that data center going, you can't stop it. You can't turn it off too much hangs on it. Oh yeah, we can we can go through that question that what you just said it can be a podcast on its own. If we're going through what you can miss. If something bad happens, but that's interesting what you said because I remember visiting Stuttgart in Germany in 2015. And the company called PTC and they were doing digital twins. So only those only then those guys were starting to look into digital twins. So if an industry like that is starting at that point, I can imagine the data center space to be maybe slightly a little bit behind around that time. So I guess, yeah, so it's like the last seven, eight years. They have really kicked in. But before we go into more of what future does. Let's maybe talk about the notion of digital twin I mean for people that don't know they're watching us. What is digital twin and what are the benefits for the data center space, especially around sustainability because there's such a big topic. So a digital twin really simple concept. Actually, it sounds, it sounds like something that maybe is complicated or something but actually, it's not really the concept of digital twin is simply that you use a model to be able to reflect the behavior of the system. So you know the obvious digital twins were like jet aircraft engines understanding when there was likely to be failure of turbine blades or anything like that. But but the principle of the digital twin can be applied to anything. And the good thing about it is that people understand the concept of a twin is really as simple as that what you're trying to do is get something that will behave exactly like it's, it's physical. instantiation it's physical brother or sister. And in this case, a digital twin is a digital twin of the behavior of the data center as you try and build it out. So it offers a great opportunity because I made the comment earlier that the one thing you know about a date center is that the design will probably never be seen in practice. And that presents right from the beginning of the lifecycle, the, the designer, a problem, because how do they try and account for the fact that the data sense can be used in a way that's a little different to the way that they imagine in their concept design. And in a digital twin you can try different things and see what happens, and therefore be able to try and avoid scenarios that are going to undermine the data center. But the beauty of it is once you've got that twin, it can grow it can involve it can, it can mimic all the changes that occur in the data center so the challenge becomes, how do you get your workflows to be reflected continuously in the digital twin so it remains a twin and doesn't become the ugly sibling. I like that doesn't become the ugly sibling. But it's a very interesting technology I mean, I actually I saw your technology a few years ago so about 780 years ago. And it was the first time I actually put a VR sets on as well. It was with you. And it's it's very interesting because you can see all the air flows you can see cooling you can see everything you can see the real thing at the real time. What's being waited and what isn't. So there's a massive play into sustainability front. Are you seeing a lot of customers really making big improvements by using digital twins and simulation software. The whole reason for doing this is to make data centers efficient and sustainable that there's, there's, there's really no point in doing that because that plays into the environmental issue which has become so politically important and for each one of us very very important for our whole being in the future, but it underpins the ability to be able to maximize the efficiency and effectiveness of the business. So the business that's operating. If they want to be sustainable, which everybody does nowadays, you need to be able to make sure that you're using efficiency you're using water, you're using all these things really effectively. And a digital twin enables you to understand exactly the consequences of any changes that you might make all the design you you create. In terms of not only will the computer overheat, but how much energy am I going to use how much water am I going to use to be able to keep this system up and running. So, yeah, huge impacts, even prior to the current round of data sense and when we first started, you could be talking about a saving of, well, not even just a saving, but there's a capacity limit so you design a data for say 10, 10 megawatts. And when you get to five or six megawatts, you start to see hotspots that are preventing you deploying equipment. So if you don't manage that, then you've wasted a lot of that investment which is material again it's sustainability because you put a lot of material into that data center you're expecting this level of productivity, but you can only get 60% of it. And so people then build another data center. That's terrible use of the earth resources. And it's obviously wasting energy because you're having to over cool it almost certainly to avoid the problems that you're, you're, you're facing. So, even in the early days, when people were just thinking about it from the design point of view we do troubleshooting for existing operators, and they'd be stuck at this 60 70% glass ceiling of not being able to and they'd say we just want to, we just want to get that capacity back. And what you'd find is that what they would do is they would not only go for the energy efficiency that they go for increasing the capacity and then just put more equipment in there. So enabled much better use of the investment we've made of the world's resources in those data centers. And if you look at that over the period of time that we're talking about, you know, the power usage effectiveness has gone from, you know, the actual values might be 2.2.0 2.5. Right. But now we're talking about best in class data centers of only being say 1.1. So 10% of the energy used is used for the infrastructure and the, the one the 100% out of the 110% is going to the it and actually doing work. It's a massive improvement so instead of previously it was like 40% of the total now it's nearly, you know, it's 90% plus of the total. That's what advanced in technology, plus the ability to understand how to put these things together using digital systems and how to continue to operate them has delivered in terms of the environment so data centers have have been incredibly effective at learning how to be a responsible player in the world environment. In some aspects we are leading by example, and kind of setting a scene as well for the industries around sustainability but that's the day is very interesting I mean I haven't really heard of a 2.5 PUE data center before. So I guess it's this well past my time. Yeah, yeah. I remember 1.6 and that was already quite bad at the time so yeah 1.1 is what we want now even less than that if possible. Talk us through the future facilities itself though because the company has grown a lot even since the last time I went to office. It's it's grown a lot there's been an acquisition, there's been quite a lot of things happening in the middle. Give us a rundown of where future is today. So, future facilities was set up really to look at this, this problem of the life cycle of the data center and understand how to manage it from an efficiency and sustainability point of view. That that goal is still very much our primary goal. But what's changing is that the vision of supply chain integration has has become much more of a reality. And what you see now is that problems in managing a data center for example might be that if equipment gets a bit hot, it has its own control system and it will demand more air or more water if it's liquid cooled perhaps. But you may not have that availability for the entire facility and so that might actually be negative in terms of your performance rather than positive. So future facilities is much broader now it covers the supply chain from the IT equipment and the design of that right through to the outside world and how the external environment impacts the performance of the data center. Be it an array of chillers on the roof, or be it IECs in direct evaporative cooling units, providing air from using air from outside to create the the cooling without mechanical cooling. So we cover that whole chain and that was the vision but we started off really in just the white space in the data center, and we're expanding out to a greater and greater part of that supply chain. And as you mentioned, the acquisition, having been acquired by Keynes design systems in July, it's opened up a huge opportunity for future facilities to do many of the things we wanted to do, but much more effectively, not just because we're part of a big company but because that big company is already heavily present in the equipment design which is so intimately linked to data center performance, but also because some of the technologies that they have in the sense of machine learning. So creating reduced automobiles etc will make decision making much faster for people within the industry and we can really connect the whole supply chain from an even tinier scale that Keynes are right down inside the chips right through to the outside world so that the whole space can be optimized. Much made in heaven. I was going to ask just like housekeeping, how big is the workforce now? Okay, so future facilities was just under 100 people, about 90 plus people. So we're a reasonable company but we've joined an organization which is near 10,000 people. So you can see the opportunity now to really scale up and be able to deliver the big vision that the data center have seen us for but they know we're just a small company in this, playing with huge multinationals and hyperscalers and the like. Yeah, because it's the thing, I mean, being up on future rather than the whole group but even close to 100 people, it's a very big company but the way the market has changed 100 people, it probably doesn't answer the needs of global markets. We always knew we had to be a global company. And hence, hence why we had offices in London and in the US and in Japan because we had to be out there as well as obviously the distribution channels. Yeah, we had to be a global market but no it doesn't. And that was the one of the reasons for the need to find a partner is that to be able to meet the demands of these huge companies that you know how fast they're moving some of these big companies in terms of constructing data sense and filling them up. We just need it need to be able to grow faster than a small company can realistically do. The speed for data power. It's never fading. I was actually going to ask, and I think you've already sort of answered the question I was going to ask, what are your strongest geographies and what are the new geographies are going to go into, especially being part of a new new group. Okay. Well, geographies are really dictated to some extent by where where the active markets are so from the data center point of view, clearly we operate in areas where there are strong data center markets. So, you know, it, we have bases in the US for example in in on the West Coast in San Jose and on the East Coast in New York so the West Coast is giving us the supply chain part of it and the East Coast is giving us the, the end user market, the large end user market and the financial markets. So anywhere that you see that sort of opportunity is a great opportunity for for future facilities. Obviously, we have much more presence and business now in the Far East, because of where I'd wear manufacturing is and all that sort of sort of thing so that's driving our growth in in in the Far East. And it's going to continue that, you know, so I, I think we will become truly global in the way we operate and that that's, again, is a reason for the need for that position to be able to operate in territories directly. They're very interesting and I can't wait to see the proper expansion across Southern Europe, Southern Hemisphere. The Southern Hemisphere is where we have to focus to expand because we have less presence in the Southern Hemisphere. That's a good point. Yeah, that'll be very interesting to watch. And then Mark, so if people want to find out more about future facilities and getting touch, where could they go? Well, it's really simple. I just go to futurefacilities.com and that will give you all the links that you need to be able to find out about what we do with the digital twin and at both the data center scale and the electronic scale. Okay, very simple and very interesting technology. Mark, so thank you so much for your time for speaking to me and it's been it's been a very long time. So it's nice to see you and see that the business is doing well and the acquisition was very exciting to get the press release about it. When it happened, it was, okay, this is good. Yeah, we're really excited. We did it. You said it match made in heaven. I think we were very fortunate to find such a good partner. Yeah, and it carries the interest in a nice place at the moment where these businesses are having around from the mid 2000s and even early 2010s and are finding a new home and there's just the market dynamics and that's very interesting and nice to watch as an outsider. Thank you. But Mark, thank you so much for your time. As for our viewers, thank you for tuning into JSA TV and JSA podcasts and don't forget to check out social channels for more content. Until next time, happy networking.