Over The Edge

Supporting Energy Supply, Reducing Carbon Emissions with David Holmes, General Manager of Energy at Dell Technologies

Episode Summary

This episode of Over the Edge features an interview between Matt Trifiro and David Holmes, GM of Energy at Dell Technologies. David is involved in advancing decarbonization and the transition to sustainable energy. In this episode, Matt and David discuss the challenges of increasing carbon-neutral power, rising energy demands, and how edge computing and technology can help improve sustainability.

Episode Notes

This episode of Over the Edge features an interview between Matt Trifiro and David Holmes, GM of Energy at Dell Technologies. David is involved in advancing decarbonization and the transition to sustainable energy. He works to understand the problems that energy companies are facing, so that operators and regulators can be more aligned when reducing their carbon footprints. 

In this episode, Matt and David discuss the challenges of a global increase in energy consumption. David explains the need to advance energy systems and technology in order to reduce CO2 emissions. He dives into the importance of making energy not only sustainable, but abundant, affordable, and reliable. David also details how edge computing and use of certain technologies can help improve sustainability.

---------

Key Quotes:

“As demand for energy is dramatically growing, we need to completely transform our energy systems to reduce CO2 emissions on a massive scale. And it's that tension that gets me out of bed every morning - understanding the role of technology in supporting the growth of supply of energy, whilst at the same time dramatically reducing carbon emissions.”

“For me, a lot of what edge computing is really about is thinking: “How can we bring the best capabilities of operational technologists and informational technologists together to build a contemporary IT environment that supports the needs of the energy industry for national, critical infrastructure.”

“It's the wild frontier of managing our energy systems, and solving some of these problems is going to be critical. But guess what? None of it is going to work without a huge amount of edge technology deployed at scale with much greater computational capability than is available today.”

“There's two problems that the OT industry faces: reliability and resilience, which are absolutely critical. They have to keep the lights on. And in the current world in which we live, that means cyber resilience becomes one of the most important challenges that any energy company has to address.”

“There is going to be compute and data across the entire continuum, right from that little valve all the way up to a hyperscale data center. And the people who are going to be successful are the people who work out the right technical and economic places to run all of these different workloads.”

“I’ve got three goals. Number one, accelerate the energy transition. Number two, advance decarbonization. And number three, ensure global energy security. And I have the entire resources of Dell's engineering portfolio to help me build the solutions that are going to let me do those three things.”

---------

Show Timestamps:

02:12 - David’s technology origin story

06:58 - The increasing demand for energy

08:08 - The climate consequences of technology

11:07 - Applying technology to help reach sustainability goals

19:47 - Edge computing in the power industry

22:07 - The intersection of IT and OT

24:02 - Challenges the OT industry faces

24:48 - Taking a contemporary approach to edge computing

33:18 - Aligning operators and regulators

35:58 - The wild frontier of energy systems

38:14 - Adopting realistic cloud computing principles

40:09 - David’s team’s goals

47:26 - Defining where the edge is

52:05 - The future of computing

53:22 - David’s daily motivation

--------

Sponsor:

Over the Edge is brought to you by Dell Technologies to unlock the potential of your infrastructure with edge solutions. From hardware and software to data and operations, across your entire multi-cloud environment, we’re here to help you simplify your edge so you can generate more value. Learn more by visiting DellTechnologies.com/edgefor more information or click on the link in the show notes.

--------

Links:

Connect with Matt on LinkedIn

Connect with David on LinkedIn

Learn More: 

www.CaspianStudios.com

Episode Transcription

[00:00:00] Narrator 1: Hello, and welcome to Over the Edge.

This episode features an interview between Matt Trifiro and David Holmes, GM of Energy at Dell Technologies. David is an advocate for decarbonization and the transition to sustainable energy. He works to understand the problems that energy companies are facing, so they can reduce their carbon footprint.

In this episode, Matt and David discuss the challenges of a global increase in energy consumption. David explains the need to advance energy systems and technology in order to reduce CO2 emissions. He dives into the importance of making energy not only sustainable, but abundant, affordable, and reliable. David also details how edge computing and use of certain technologies can help improve sustainability.

But before we get into it, here’s a brief word from our sponsors.

[00:00:50] Narrator 2: Over the Edge is brought to you by Dell Technologies to unlock the potential of your infrastructure with edge solutions. From hardware and software to data and operations, across your entire multi-cloud environment, we’re here to help you simplify your edge so you can generate more value. Learn more by visiting Dell.com for more information or click on the link in the show notes.

[00:01:10] Matt Trifiro: Two years ago when I started the Over the Edge podcast, it was all about edge computing. That's all anybody could talk about. But since then I've realized the edge is part of a much larger red.

Pollution. [00:01:40] That's why I'm pretty proud to be one of the founding leaders of a non-profit organization called the Open Grid Alliance for oga. The OGA is all about incorporating the best of edge technologies across the entire spectrum of connectivity. From the centralized data center to the end use devices, the open grid will span the globe and it will prove performance and economics of new services like private, 5G and smart retail.

If you want to be part of the open grid movement, I suggest you start@opengridalliance.org where you can download the original open grid manifesto and learn about the organization's recent projects and activities, including the launch of its first innovation zone in Las Vegas, Nevada. 

[00:01:55] Narrator 1: And now, please enjoy this interview between Matt Trifiro and David Holmes. GM of Energy at Dell Technologies.

[00:02:03] David Holmes: So David, how are you today? Good, thank you. Nice to be here. Thank you 

[00:02:06] Matt Trifiro: for joining us on the podcast. You know, one of the things I, I like to start with is just asking my guests how you even got into technology. Was there some forming moment? 

[00:02:15] David Holmes: Yeah. My origin story is quite fun. When I was eight years old, my father bought [00:02:20] home an OLiveti P 60 60 mini computer.

[00:02:23] David Holmes: It had twin eight inch floppy disc, drives, a single line L C D display and a thermal printer, 48 kilobytes of ram and a proprietary process. And I sit and programmed a Star Wars game in basic on this computer, and that's kind of how I got into computer. 

[00:02:40] Matt Trifiro: Wait, is it, isn't Olivetti 

[00:02:41] David Holmes: an Italian company? It is.

[00:02:43] David Holmes: It's an Italian company. Yeah. I remember the 

[00:02:45] Matt Trifiro: TS 80 and I remember the Apple two and I remember, was it that era? Was it like the This is, 

[00:02:49] David Holmes: this is way before that. So the Oliveri P 60 60 was then replaced by a Commodore pet. Commod Pet was then replaced by a Commodore 64, which is. The, the era of computing and technology that you might actually have have heard.

[00:03:03] David Holmes: That's 

[00:03:03] Matt Trifiro: interesting. Yeah. Wow. Okay, so you wrote the Star Wars game, so it was after Star Wars. 

[00:03:08] David Holmes: Well, I said, I wrote the Star Wars game. I cried out of a magazine in the olden days, magazines used to come with 

[00:03:14] Matt Trifiro: 30 pages of code. You could type in . Wait, we were dedicated back then, weren't we? It seems crazy.

[00:03:20] Matt Trifiro: Okay, so you fancied yourself a young programmer. Yeah. So when you type that all. 

[00:03:25] David Holmes: Yes, absolutely. It was the craziest game because you'd basically type in a [00:03:30] velocity and a heading and it would sit there for a minute, and then the little thermal printer would print out 20 lines of what happened, and then you'd adjust your, you'd play it on the thermal printer.

[00:03:40] David Holmes: Yeah. You'd, you'd then type in a new velocity and a new heading. I love it. It would then tell you whether you were on track, off track, and in each run of the game would be like, I don't know, 10 feet of thermal. That 

[00:03:51] Matt Trifiro: is, that is really awesome. When I was young, my father who was in the insurance industry brought home a calculator.

[00:03:57] Matt Trifiro: This was Charles Babbage style mechanical calculator. I mean, it had it emoted and you'd push a key and a bunch of gears and clacking it would print on the paper. So you are more sophisticated than I than I was, and you know, now your specialty is, Energy. That's right. Can you connect us between programming Star Wars out of thermal paper and your sort of current passion?

[00:04:18] David Holmes: Wow. It take me a couple of hours to tell you the full story. G gimme, gimme me the two or three highlights. I went to college in the uk, did computer science from there, went to work for a, a PC manufacturer called research machines, worked in computer aid design from. Uh, in 1993, Microsoft launched this exciting new operating system called Windows nt.

[00:04:39] David Holmes: Not, 

[00:04:39] Matt Trifiro: not [00:04:40] everybody would've called that 

[00:04:40] David Holmes: exciting. Well, it was exciting for me and I was, I was the 54th Microsoft certified professional in Windows NT in the uk. Um, so very early into that, and that led me into a role in enterprise architecture, which took me up to Scotland, where I worked for the Royal Bank of Scotland.

[00:04:56] David Holmes: Did a lot of their distributed architecture. And then like so many stories in our lives, I met this girl and decided I wanted to marry her and I needed to move back to England, and I got this three month contract working for this oil and gas technology company. And it was really just a temporary thing, but I learned all about.

[00:05:15] David Holmes: Some of the data types associated with oil and gas, seismic data. Well data, how reservoir characterization works in the intersection of it on oil and gas. And that company then got acquired by Halliburton and I spent 10 years there where I ran there. This was your temporary job, ? Yeah. This is my three month job.

[00:05:34] David Holmes: 10 years later. I was running there, amea IT and hosting operations. From there I went to do a startup for five years, which built some cloud-based data management platform. And then I got a phone call from EMC saying, Hey, we'd like a CTO for our energy [00:05:50] business. Are you interested? And it was just the point that the startup was being acquired.

[00:05:54] David Holmes: So I'd I leapt into that and. Last year got promoted to be general manager of Dell Technologies energy business where I've been for the last 15 months. And so I now run a global energy business and lead a team of CTOs, field directors, solution specialists, energy sustainability consultants, and others in developing our energy business.

[00:06:17] David Holmes: That's interesting. When I 

[00:06:18] Matt Trifiro: think of the intersection of software, hardware, technology, Silicon Valley, the world, Dallas Square in, and I think of the world of energy, uh, I think of this as being very different. And I also think of two times of energy, and you mentioned both types. So one is serving the energy industry as a vertical.

[00:06:34] Matt Trifiro: Yeah, right. Which is how do we make mining more efficient gas and oil exploration, efficiency of the refining, I mean, all of that. And I'll let you actually want, want to hear your perspective on, on what's pressing there. But also you mentioned conservation, sustainability. Does the Dell Energy business only focus on the energy vertical, or does it also work on energy problems in other, 

[00:06:55] David Holmes: in.

[00:06:56] David Holmes: It works across the intersection of both. Here's the fundamental [00:07:00] challenge. In the next eight years, there are gonna be a billion extra people who are gonna start becoming material consumers of energy. 

[00:07:06] Matt Trifiro: How many are there today? See, there's a billion more coming. 

[00:07:08] David Holmes: Yeah, there's probably about 6 billion today.

[00:07:11] David Holmes: There's gonna be about 7 billion in in eight years time. Developing countries are developing unreasonably. The populations want access to abundant, affordable energy, but at the same time as demand for energy is dramatically. We need to completely transform our energy systems to reduce CO2 emissions on a massive scale.

[00:07:32] David Holmes: It's that tension that gets me outta bed every morning is understanding the role of technology in supporting the growth of supply of energy, whilst at the same time dramatically reducing carbon emissions. So, When we think about our overall sustainability strategy at Dell, there are really three elements to it.

[00:07:49] David Holmes: One is the things that Dell needs to do as a company to be resilient to climate change and to take accountability for how we operate our business to look at our own carbon footprint. The second is to look at the sustainability challenges associated with the use of it, depending on which report you.

[00:08:08] David Holmes: You know, it is [00:08:10] somewhere between two to 3% of all global greenhouse gas emissions that's expected to grow to four to 6% within the next five or six years. So there's a big problem to be addressed with understanding how we mitigate and abate the climate consequences of the use of technology. That though is dwarfed by the emissions associated with the energy industry, which is over two thirds.

[00:08:32] David Holmes: All emissions. So there are people with Inside Dell who are looking at all those three pillars, but mine and my team's focus is how can we apply technology? To address the two thirds of global greenhouse gas emissions associated with the energy industry. So what do you mean 

[00:08:49] Matt Trifiro: by that? You know, two to 4% is caused by it, and then you say two thirds is the energy industry.

[00:08:55] Matt Trifiro: Help me understand what you mean by two thirds of what and where's that 

[00:08:59] David Holmes: coming from? Okay. So if you think about the energy industry, think about. Power generation. Think about transportation, think about anything. The 

[00:09:07] Matt Trifiro: why is transportation part of the indu energy 

[00:09:09] David Holmes: industry, because you take fossil fuels, you put them inside your car and you burn them.

[00:09:15] David Holmes: That generates energy. So it's not just 

[00:09:16] Matt Trifiro: the production of energy, it's, well, it's 

[00:09:19] David Holmes: everything. If you look [00:09:20] at how companies think about their carbon footprint, right? They have Scope one, which is their direct emissions. It's, we burnt something. It generated co2. Yeah, that was our business. You have scope two, which is indirect.

[00:09:33] David Holmes: And indirect is kind of like, well, I got on a plane and I flew to London. That plane burnt some fuel. 

[00:09:39] Matt Trifiro: I bought somebody electricity, but it was Texas Energy that burnt the fossil fuel. 

[00:09:44] David Holmes: Exactly. But then you have Scope three, which is your downstream supply. I paid some companies to make some things for me, and in making those things, they generated co2, but then you have upstream emissions, which is the really fun part.

[00:09:58] David Holmes: Upstream emissions are, I sold you a laptop. You plug that laptop in. There were emissions associated with the electricity you used to power your laptop. Now, if you look at Dell's carbon footprint, about 98% of it is scope three of that 98%. About a third is downstream supply chain emissions, and two thirds is use of sold product.

[00:10:21] David Holmes: Split roughly 50 50 between client devices and data center devices. So if you think about a company like Dell, which has said, we will be net [00:10:30] zero across scope one, two, and three by 2050. Scope one and two are tiny, but we have to address all the emissions associated with the use of our sold products across the.

[00:10:43] David Holmes: That's the really big and complicated part that's fascinating. 

[00:10:47] Matt Trifiro: And I'd actually never realized this. And I consider myself kind of a, a fan of conservation and sustainability. Not an expert, but a fan. I never really fully absorbed that a company like Dell that has a sustainability policy isn't just looking at what they do, but what their customers do with their products.

[00:11:03] David Holmes: And that leads us on to the next bit. So on the one hand, we, Dell, we have an obligation to address. All of the emissions associated with the energy used to power our products. That's used by, that's purchased by our customers. At the same time though, coming back to my role, what I'm really interested in is how can we apply that technology to help our energy customers achieve their own climate and sustainability goals?

[00:11:29] David Holmes: Now, 

[00:11:29] Matt Trifiro: walk me through the eyes of your customers. So, I'm an energy company, pick one, oil, gas, electricity, and just, how am I thinking about, because I make energy, I guess I have a carbon f. [00:11:40] How do I think about it? How do things like the difference between oil and gas and hydroelectric and, and then how am I thinking my downstream, like everybody that uses my electricity is creating?

[00:11:49] Matt Trifiro: Let's 

[00:11:50] David Holmes: imagine that you are a multinational, super major oil and gas company, okay? And you can pick anyone you like, but if you go to their website, You will read on the front page about their climate and sustainability objectives. Now, there are differences between them. Some companies, maybe more European oriented, have set more aggressive goals.

[00:12:11] David Holmes: Some US based companies have set slightly different goals for approaching it in different ways. What you'd see is European super majors are typically looking at diversification and saying, actually, we are an energy company, so we need to rebalance our portfolio, have a much greater proportion of renewable and sustainable energy generating capacity.

[00:12:29] David Holmes: In the US I'd say companies are more focused on decarbonization. How can we address methane emissions challenges? How can we use our expertise in the subsurface to. Carbon capture utilization and storage. But wherever you are, you will have a set of climate and sustainability goals that talk to how you are gonna address your emissions.

[00:12:48] David Holmes: Now, let's come back to that scope [00:12:50] three problem that I talked about a minute ago. So Dell makes a lot of physical devices, which need electricity to run, and there are emissions associated with that. If your product is hydrocarbons, you have a much bigger scope three problem. In other words, when you sell your customer a gallon of Gasol, Then

[00:13:09] Matt Trifiro: That's all. It's all, that's all 

[00:13:10] David Holmes: hydrocarbons. There's no mitigating it. So how in the world as as a company do you, do you address that? These are the sorts of challenges that we're working on? Let me give you an example of the sort of ways that technology can address these problems. So if you think about the electric grid, one of the things that we pay a lot of attention to is something called the grid emission factor.

[00:13:32] David Holmes: The grid emission factor says how many grams of co2. Are being generated for each kilowatt hour of energy being consumed, and it changes. If it's a very sunny day and the wind's blowing really strongly, you'll have a lot of solar and a lot of wind. So your grid emission factor goes right down. If it's the middle of the night and there's no wind, then you're gonna be reliant upon much more fossil fuel based generating.

[00:13:56] David Holmes: So your grid emission factor is going to go up. Now, if you [00:14:00] look at the situation in California, They had a situation where they had to curtail or switch off a whole bunch of their solar generating capacity. Because they didn't have anywhere for the electrons to go. And then a few weeks later, they're sending out emergency messages to everyone with an electric vehicle saying, please, please, please do not charge your EV from 4:00 PM to 9:00 PM every night.

[00:14:20] David Holmes: Because guess what? They didn't have enough power. Okay. I 

[00:14:23] Matt Trifiro: wanna, I wanna interject because there's something here that you just glossed over cuz I think it's your your world. But what do you mean the electrons had no place to go? What does that even mean, ? 

[00:14:33] David Holmes: Well, it's really simple. The electric grid was designed in 1892 by Thomas Edison and he had this brilliant design.

[00:14:40] David Holmes: It basically said, we're gonna build some giant power stations, which are gonna generate huge amounts of electricity. That electricity will go over a transmission network. That's the kind of the big pylons, the steel pylons that will go to some substations, which will then step down the voltage and send the power over a distribution network.

[00:15:01] David Holmes: That's the kind of the, the wooden poles and wires that will then go into homes, businesses, and industry as needed. And the grid has always [00:15:10] been operated on a supply management. People got really, really good at predicting how much energy we needed and then said, okay, we just need to make sure that we're generating enough energy plus a small margin to supply everyone's power needs.

[00:15:25] David Holmes: You know, you used to have things like, Hey, it's the Super Bowl this weekend we're gonna have an issue generating enough power. . So what we're gonna do, we're gonna call this aluminum plant over here. We'll give them a couple of months notice and say, Hey, can you shut down on Super Bowl day? Because you know, the grid's gonna be under a bit of load and, and, and that'll be great.

[00:15:44] David Holmes: But it was really fundamentally SupplySide manage this. We'll figure out how to make as much energy as needed. Now all of the generating technologies that Thomas Edison were aware of were basically 24 by seven. You pour in coal, you pour in gas, you burn oil and power is generated. Nuclear exactly the same.

[00:16:03] David Holmes: Well, what happened when we started adding wind and solar? Well, all of a sudden some of our generating capacity is not dispatchable. We can't decide when we're gonna turn on a wind turbine and when solar, solar powers gonna be generated, it's intermittent. We can predict when it will [00:16:20] generate, but we can't choose when it's gonna generate energy.

[00:16:24] David Holmes: So when there's very small amounts of intermittent generating capacity, we can balance the grid and we can manage things really. But now we're seeing intermittent generating capacity as 20, 30, 40, 50% of the generating capacity, and it's intermittent. So now we have the situation where the sun's shining really brightly in California.

[00:16:45] David Holmes: The solar's coming on stream. It's generating huge amounts of power. If there is not the demand to consume that power, you've gotta do one of two things. Well, one of three things I guess. You've either gotta create. So you could, for example, go to some industrial freezer storage facility and say, Hey, you are running a kinda like minus two degrees Fahrenheit right now.

[00:17:08] David Holmes: We'd like you to crank up your chillers. Take a huge amount of 

[00:17:10] Matt Trifiro: power. That's really interesting to use the refrigeration company as a, as a battery e. Exactly. store that cold energy. That's 

[00:17:18] David Holmes: fascinating. You can do that. You can use utility scale battery storage to achieve the same thing. You either create load on the grid, you store energy in a battery, or you shut down [00:17:30] generation.

[00:17:30] David Holmes: The problem is because the grid's always been designed, we've worked like this for the last 130. And it hasn't transformed in the way that we need it to because what we need is a grid that can dynamically manage the demand side as well as the supply side. So here's my crazy silly example that I always use.

[00:17:49] David Holmes: I have a smart dishwasher. It's grid enabled, okay? Because when I go to bed at night, I set my dishwasher going. I don't care when it runs. I wanna wake up in the morning and have clean dishes, but I don't care at what point during the night it. I've got an electric vehicle. I plug it in at night. I don't actually care when it charges.

[00:18:07] David Holmes: It can charge anytime, but I'd like to wake up with enough power so that my daughter can drive herself and her brother to school in the morning. All of a sudden, we have huge amounts of load, which we can schedule. What we can do with technology is rethink the way that we operate The. because grids are typically operated by these huge control rooms with kind of 15, 20 people in them and, and they monitor and manage the grid and deal with outages and availability.

[00:18:34] David Holmes: But here's the thing, there are never gonna be enough people in that control room to sit down one evening and go, [00:18:40] okay, Mr. Holmes has set his dishwasher going and he needs his EV charging. What time are we gonna do that? And how are we gonna optimize those jobs for minimum carbon intensity of consumed power?

[00:18:50] David Holmes: Now, Mrs. Jones, she set her washing machine going, and Mr. Frederickson, he's turned his ice maker onto to make me lots of ice, please. So let's figure out how we're gonna schedule that in all of. Has to be completely automated using autonomous systems that work right at the very edge of our networks because not only do we have the complexity of all of this utility scale, intermittent generating capacity, but now we've got consumers with their own solar, with their own EVs.

[00:19:23] David Holmes: All of a sudden they're generating and consuming energy and providing a storage resource as well, and it's gonna take. Order of magnitude, more compute capability, right at the very edge of our energy networks, to be able to autonomously manage these, you know, hundreds, thousands of times, more devices in order to minimize the carbon intensity of consumed energy and make sure that we're balancing demand with supply.

[00:19:49] Matt Trifiro: All right, so let's talk [00:19:50] about. Edge computing as it applies to the, to the power industry. Um, I'm gonna unpack some of what you said just so that our audience is on the same page. So you talked about supply and demand. The supply, as you mentioned, is anything that generates power. And we've got all these variety of ways of generating power at large scale.

[00:20:05] Matt Trifiro: We've got wind turbines and hydroelectric dams and nuclear power plants and all these things on the traditional demand side. We also have this other emerging demand side, which is like, I've got solar panels on my roof and maybe there's other kind. Power generation that happened in distributed fashion that can be contributed to the grid.

[00:20:21] Matt Trifiro: And then you have the, the demand side. And the demand side is like when I term on my thermostat, I create demand in my house and I'm consuming electricity. And so what you're saying is that by applying computer technology, by applying algorithms and automation and networking and all of these things, we can connect all of.

[00:20:38] Matt Trifiro: Devices together and in, in essentially real time rationalize like what are the highest and best uses of the existing system for generating and suppressing electricity, which I actually think is maybe the most interesting part of this in a way that that isn't obvious. Because if you think about peak load, right?

[00:20:57] Matt Trifiro: That's what everybody's got their air conditioners on. You [00:21:00] have to build an electrical system that's capable of delivering enough power with peak load or you need a way to curtail the. Right. And when it was at Human Pace, right, you could call the aluminum factory up or the, you know, the freezer 

[00:21:13] David Holmes: company up.

[00:21:13] David Holmes: And that's what a lot of people are focused on is give me energy that I can dispatch on demand. But the real challenge to your point, and we saw this in Texas when Winter Storm Yuri hit, is the technology that was available. To curtail loads was brutal. It's a thing called load shedding where you shut off whole communities and neighborhoods.

[00:21:35] David Holmes: And the consequence of that was hundreds of people died because we didn't have the technology to better manage the, 

[00:21:44] Matt Trifiro: but imagine if you could turn down the thermostat of a million homes by 

[00:21:48] David Holmes: one degree. Yeah. And that's a really tough thing. You know, I, I, I, I live in Texas. I went through. It's pretty frustrating knowing that, you know, as a society we weren't lacking the technology that's needed to be able to solve this problem.

[00:22:01] David Holmes: Yeah, 

[00:22:01] Matt Trifiro: that's right. Okay, so let's, let's talk about what is the technology that is available to solve this problem and what's stopping it from being 

[00:22:07] David Holmes: solved? This is where you run [00:22:10] into the intersection of. Information technology and operational technology, and that's a really interesting place where edge computing is really sitting because for me, a lot of what edge computing is really about is thinking how can we bring the best capabilities of operational technologists and informational technologists together?

[00:22:32] David Holmes: To build a contemporary IT environment that supports the needs, you know, in my industry of, of what is national critical infrastructure, and there's always been a tension between IT organizations and OT organizations. The cliche is the OT guys sit there and go, you it guys, you good with email and internet access, but if anything you run goes down, there's really, it's no big.

[00:22:56] David Holmes: you know, we run plants and the electrical grid, and if we get it wrong, then really, really bad things happen. and the IT guys sit there going, you guys are working 20 years in the past. You've got like these Windows XP systems running this critical workload that can't possibly be protected and you refuse to have anything upgraded or have patches applied more than every three years.

[00:23:19] David Holmes: There's [00:23:20] been a real tension between those organizations, and it's one that I think edge computing really starts to address. So if you take a typical substation, for example, if you go into a typical substation, you'll maybe find 13 to 15 what are called single function devices. This is a box that looks to you and me very much like a, a pc.

[00:23:42] David Holmes: You know, there's a process of some memory and some storage and some networking. And it does one thing. It performs one function. And if you look across all of these devices, they'll run different operating systems. They'll run software from different vendors. So you're 

[00:23:55] Matt Trifiro: trying to tell me they're not built on cloud principles.

[00:23:57] Matt Trifiro: It's 

[00:23:58] David Holmes: not always the most contemporary software stack. But the, but the challenge is this, that, you know, there, there's two problems that OT industry faces. One is reliability and resilience is absolutely critical. They have to keep the lights on and. in the current world in which we live, that means that cyber resilience becomes one of the most important challenges that any energy company has to address.

[00:24:24] David Holmes: So that's the first problem. How do I keep it all running and how do I keep it all secure? [00:24:30] And if you've got 500 substations and each of them has got this random collection of, of devices, then there are some real challenges. I don't mean to be kind of panic mongering, but you know there there's a huge amount of people doing a lot of.

[00:24:43] David Holmes: To ensure that our infrastructure is secure and protected. It's a very strong regulatory environment. But what we can say is that as we look forwards, a contemporary approach to edge computing is going to solve for that problem of building a robust cyber resilient environment and solve for the second problem, which is we need to deploy orders of magnitude, more compute and data capture capability.

[00:25:08] David Holmes: Right at the edge of the energy networks to be able to deploy these new autonomous workloads that are gonna help drive down the carbon intensity of consumed power. . 

[00:25:18] Matt Trifiro: That's really interesting. And you know, it's, it, there's a direct analogy. So I'd know a tiny bit about the power industry, and I'm learning more today, but I know a lot about the telco industry and the, the analogies I imagine are not just between those two industries.

[00:25:31] Matt Trifiro: I mean, you walk into, you know, go to a traditional cell tower and at the bottom of the cell tower is a base station that is a dedicated appliance. It's exactly what you're saying. You open it up, it's probably a Windows [00:25:40] machine, right? Like literally there's a, there's a, there's a board in there that's running windows and you know, all the bank ATMs.

[00:25:45] Matt Trifiro: It's the same thing. This is, uh, something that I, that I've been thinking about quite a bit, which is you mentioned the data capture, as you say, we, we have all of the technology to understand what's happening in a complex system like a grid. And to reason around it. To even build a digital twin, they can run AI models against the the physical infrastructure.

[00:26:04] Matt Trifiro: But what seems to be lacking is the high fidelity data of all the components to do this. And we see this in the telecom industry. We see this in the computer industry. Now, the software industry has solved this to some extent, or at least they're the most advanced because they've moved from telemetry and analytics to observability, observability.

[00:26:23] Matt Trifiro: A built-in characteristic. It is a characteristic of a system that it is observable. So as opposed to this, this thing like, oh my God, we got the system. We have to go figure out how to measure it. Let's go put probes in and sensors. It's like, no, no. You have a system. And that system's been designed so it spits off the data that you need to be able to reason about it.

[00:26:39] Matt Trifiro: And I think it's a fundamental difference, this very early path that we're, that we could be on, which is to build observability into every single. [00:26:50] Physical device thing we put out in the field, whether it's a transformer or a network box or a server, it should know how to generate its own observability data.

[00:26:59] Matt Trifiro: Tell me about the data collection, the observability, and how we get that to happen at scale. 

[00:27:03] David Holmes: There's a lot that I could talk about because you've hit exactly the kind of the, the area that we spend a lot of time looking at. So, Let's pick on one example. Smart Meters is a device that is able to capture and store energy usage information and transmit it over a network to some data capturing device.

[00:27:25] David Holmes: So in the 

[00:27:26] Matt Trifiro: old days, I had the spinning thing and the dials and the guy had to come out and read it. This is one of those, but it just sends back data on this is how much energy's being used in this location. A 

[00:27:34] David Holmes: absolutely, and people called these things smart meters. Because they replaced the need for somebody to turn up at your house with a book and write down a number and then go back to the office and type that number into a computer, which would generate a bill.

[00:27:50] David Holmes: Most smart meters have less compute power than that. 65 Oh Intel 65 0 2 Pad Commodore Pet that I used however many decades ago. They capture something like [00:28:00] 56 bites of data every 15. And you talked about fidelity, so that usage data is incredibly helpful from a billing perspective. because now what you can do is you can start being very flexible about the way you, you build billing contracts.

[00:28:16] David Holmes: You can charge people different rates a day or at night, or you can set special incentives. I purchase my energy on the wholesale rate in Texas, which changes every five minutes. So the amount I pay changes. And so what I can do is I can watch the real time wholesale price of energy, and I can go and change my load.

[00:28:33] David Holmes: I can go, well, I'm gonna run. Washing machine at 2:00 AM because I can see that it's gonna be, or you could set a 

[00:28:39] Matt Trifiro: rule that does it. I'm not gonna go look at the rates. I'm gonna say, don't run my washing machine at, unless it costs me less than $2. There's 

[00:28:45] David Holmes: the number of it, right? Is that the, the automation to automate these things just isn't quite there yet.

[00:28:51] David Holmes: You have to be a little bit of a geek to set this stuff and make it work. And one of the big challenges is making. Basically bringing that kind of app experience to power management and automation, and it's very much coming. There are some people kind of at the forefront of this, but we're gonna see a lot of development in, in this space.

[00:29:09] David Holmes: But let's, let's come [00:29:10] back to these smart meters. They're just a data capture device. They provide very low fidelity data. Well, what would happen if, instead of capturing that usage data every 15 minutes, what if you captured it at 32 kilohertz? If you sampled 32,000 times a. Well, it turns out what you can do with that fidelity is you can actually identify exactly which devices are switched on at any one time.

[00:29:34] David Holmes: You can look at the profile and say, okay, right now the pool pump is on. AC Unit one is on just 

[00:29:41] Matt Trifiro: by the load. That is cool. That is cool. Yeah. 

[00:29:44] David Holmes: And now that raises a really interesting privacy issue. So if I'm a utility company and I put in a smart meter that has that level of granularity, all of a sudden I have a bunch of data about your lifestyle, about your habits, about what you're doing.

[00:30:00] David Holmes: The, in some ways goes way beyond everything that we freely give away to Google and. You know, really detailed, intimate information about how you live your life when you watch tv, when you cook meals, what time you leave the house in the morning, what time you come back. And so there are challenges there.

[00:30:17] David Holmes: But with all of that data, you [00:30:20] have the opportunity. To be able to transform the way that you operate the energy systems and the two dimensions of this, first of all, exactly like you said, data observability, but then autonomous compute right at the edge of the network. I need a home energy controller in my house that automates.

[00:30:40] David Holmes: All of these processes and goes, okay, I'm gonna pre-call the house overnight. Hey, turns out sometimes the energy prices in Texas go negative. So if at 4:00 AM the price of energy is minus one, then crank up the ac, get the house as cold as you can possibly get it, you're being paid to chill the house, and so on and so forth.

[00:30:58] David Holmes: So there's a, there's a huge amount that can be done to really change the way, and if you think about the goals to get to net zero, The next three decades, and if you think about the goals in the US to have a zero carbon energy grid by 2035, that's not gonna be achieved by hard technology alone. No number of wind turbines and solar arrays and nuclear power stations and hydroelectric are gonna get us to the zero carbon grid Technology is gonna be [00:31:30] absolutely.

[00:31:30] David Holmes: At the center of helping make this a reality. Yeah, 

[00:31:34] Matt Trifiro: that's, that's pretty cool. And one of the things that, that I'm very involved in, and Dell is also involved in, for better or for worse, also called the grid. It's by analogy, not literally power, but it was analogy that Sun Microsystems came up. The original cloud computing was called utility computing or grid computing.

[00:31:50] Matt Trifiro: And um, we have a version of it with Amazon and Google and, and all these, these systems. But the metaphor was actually pretty powerful. The idea. Like computing should be everywhere, just like electricity seems like. And you should be able to plug a device into the wall or connect it to a network and it should be able to get all the compute that it needs at that moment, more or less.

[00:32:08] Matt Trifiro: Giving the Yep. Ability to create the supply, to match the, the demand. And this is actually being built today. There's an organization called the Open Grid Alliance that, I mean, this really is. What edge computing needs to be cuz it's, it's very different to go and stick a device out in the field than to have thousands or millions of devices that are actually pre networked and coordinated and available as a platform that can unleash the creativity of people.

[00:32:32] Matt Trifiro: Cuz the, one of the challenges, these closed systems, and I mean these appliances and things is like there's a very small team at a private company with a very limited set of [00:32:40] incentives and resources and so on. are entirely responsible for making that thing better as opposed to the internet where everybody gets a chance to make it better, and I think it evolves much more quickly.

[00:32:50] Matt Trifiro: And so we're building this, this ability of the power grid and the compute grid because they are both operating, they're using each other. Yeah. The ability to reason about the compute grid requires power that's supplying it. And it's almost like the two most energy intensive things that we have in our world.

[00:33:09] Matt Trifiro: Computers and telecommunication networking equipment that consumes the electricity and. Electric grid that supplies the electricity. Yeah. 

[00:33:16] David Holmes: And if you take the US an example, if you look at the infrastructure investment in Jobs Act, if you look at the infrastructure, uh, the Inflation Reduction Act, there's a huge amount to support broadband networking for electric utilities, both to transform their own networks, to enable enabled, uh, communications networks that they'll need for orders of magnitude more data, but also as a way of solving some of the rural broadband network challenges of bringing equity.

[00:33:43] David Holmes: To network access. But I want to pick up on your point about where everyone can be a participant and [00:33:50] one of the real challenges in making what I've been talking about, a reality is you have power generation companies, you have grid operators. You have regulators, both, and in the US at the state level, at the federal level, not always completely aligned.

[00:34:05] David Holmes: You have retail energy providers who sell energy, and then you have a bunch of companies that are trying to disrupt the market. By creating these kind of virtual power generation and dispatchable energy capabilities. And what's gonna be really interesting is understanding how all of these organizations work together and who's really gonna be the winner.

[00:34:26] David Holmes: Ultimately, it really wasn't kinda like at and t and Comcast that were the real winners of the internet broadband world. It was Google and Amazon and all the companies that were able to leverage the capabilities that the the internet brought. They were the people who made a trillion. The telcos have done very nicely out of it.

[00:34:45] David Holmes: When you look at how the energy markets are likely to be disrupted, there's a real possibility that actually the companies that own all of the physical infrastructure will be less able to monetize the transformation of the grids. If other [00:35:00] disruptive companies come along and say, actually, we are gonna figure out how to aggregate all of these disparate resources, and again, that's silly example, right?

[00:35:07] David Holmes: Port are gonna sell like 150,000 of their F-150 Lightning. that's gonna put more energy storage capacity on the US grid than the sum of all utility scale storage that's been deployed today. 

[00:35:20] Matt Trifiro: Yes, Ford. Ford could partner with all its car owners and become a battery supplier to the grid and help pay 

[00:35:27] David Holmes: your lease.

[00:35:28] David Holmes: Absolutely right with megawatts of dispatchable energy. And so the question then becomes, well, what's the company that's going to aggregate that capability together that's going to create the interfaces with the grid operators? And who gets to monetize it? Do I as an owner of F-150 get any money back?

[00:35:45] David Holmes: Do I get a discount on my lease? Is it free because I'm making this available? Like how is this gonna work? And there, and there are, There are already some case studies of partnerships and and investigations being done in this space, but it's the wild frontier of managing our energy systems and solving for some of these problems are gonna be critical.

[00:36:03] David Holmes: But guess what? None of it is gonna work without a huge amount of edge technology deployed [00:36:10] at scale with much greater computational capability than is available today. Yeah, and 

[00:36:15] Matt Trifiro: I keep coming back to the IT OT thing because you hear that a lot. And, and I had two thoughts and I just kind of wanna run 'em by and get your, your reaction.

[00:36:23] Matt Trifiro: So the first one is, you were talking about the old schoolers in, in these large energy generation facilities, and they're like, you know what? I gotta protect my piece of iron cuz it has to run at 30 nines. And I don't trust your little, like my laptop crashed all the time. I don't trust your thing.

[00:36:38] Matt Trifiro: Outside of like the human life consequences that happened in Texas, which all those systems didn't seem to fix. Outside of the, the life consequences, but like the financial consequences, the financial consequences that a Google or an Amazon faces with a catastrophic failure are at least as bad as the financial consequences that a power company faces with a catastrophic failure.

[00:36:58] Matt Trifiro: The clever capabilities that the Amazons and Googles and other hyperscalers have developed to maintain. Lots of resilience. Seem like they're directly politic to power. So high availability I think is a, is a great example. You can build redundancy in a data center. You know this, right? By having all this mechanical redundancy, you can have five servers.

[00:37:19] Matt Trifiro: [00:37:20] You can that, that you can hot swap at any point. They're all running the same thing. You can have, you know, two backup batteries. You can have two shore powered connections. You can have three generators, like can have all these levels of mechanical redundancy in the irony as you're talking about wasted energy, right?

[00:37:34] Matt Trifiro: The ironies, those you hope you never. , you hope you never turn that third generator on or that second generator, whereas high availability, which requires a ton of data getting back to observability. But if you can look at your data center or your power structure through a digital twin, right, you can predict what things are gonna happen and what things are gonna go right and what things are gonna go wrong, and you can build enough extra capacity that's pooled to be able to fail over with very, very.

[00:38:01] Matt Trifiro: Degrees of confidence. Do you see these sort of hyperscaler techniques making their way into this old school mechanical, like, I don't trust it. I can't kick it 

[00:38:10] David Holmes: world. Not, not exactly. One of the principles upon which cloud computing is built is that everything is ephemeral. , right? Everything can go away and not exist, and we can carry on and be robust and resilient when something disappears.

[00:38:25] David Holmes: And if you talk to a P grid engineer and say, well, let's just imagine [00:38:30] that this asset here is not available and can't be recreated, that becomes highly problematic because here, here's the real challenge, right? People have come to expect abundant, affordable, reliable. And what the government has said is, in addition to abundant, affordable, reliable, resilient energy, it needs to be zero.

[00:38:53] David Holmes: and it needs to be transitioned by 2035. And by the way, it needs to be a just transition as well. So we need to take into account the social consequences of all these things as well. When you talk about kinda like are there parallels with cloud computing with public cloud providers and hyperscalers?

[00:39:09] David Holmes: Yes and no because what we're talking about is a much more distributed fragment. Environment where reliability and resilience is fundamentally different. You can't deliver power to a different neighborhood and say, well, this neighborhood's down, but that's okay cuz we're powering this one over here now.

[00:39:29] Matt Trifiro: Yeah. I, I see the analogy breaks down at some point. That, that does make sense. You know, I'd like to talk a little bit more about what Dell is doing. So specifically, I, I don't yet understand, sir, what [00:39:40] your division sell. Today and what you wanna be selling in the future. So can you like, break it down? Like what do you actually, like, what is it that you sell?

[00:39:47] Matt Trifiro: What's your product line? 

[00:39:48] David Holmes: Oh, well now you sound like, uh, one of the energy executives that I meet on a regular basis who, who starts off by saying, why is the laptop boy here? 

[00:39:56] Matt Trifiro: Yeah. Okay. Why is the laptop boy 

[00:39:58] David Holmes: here? And that's the thing because the, what's interesting is I know there's 30,000 odd people inside Dell who are really good at selling it to it.

[00:40:07] David Holmes: the job of my team is actually not to go and talk to IT people, but to go and talk to people who run energy companies, who understand the problems and challenges they're trying to face, and to understand the application of technology to solving those problems. So we've built a portfolio of products. We take our core portfolio, particularly oriented around our infrastructure, so, And we look at how we can take that technology and apply it to meet the specific needs of the energy industry.

[00:40:37] David Holmes: And we do that in through a variety of different industry validated solutions. Many times they're developed with, uh, some of our big industrial partners and ISVs. So [00:40:50] to be clear, Dell does not and has no ambition to write virtual protection relay software. We are not gonna start writing. Seismic interpretation software, or well log calibration software, but you'll build 

[00:41:03] Matt Trifiro: servers that run that really 

[00:41:04] David Holmes: well.

[00:41:04] David Holmes: What we do do is we understand what the data types are for the particular workloads, how we can optimize the engineering of our infrastructure to best run those workloads, and perhaps more importantly, ensure that we meet all the necessary regulatory requirements. So you can't just rock up at an electrical substation with your latest computer and throw it in.

[00:41:28] David Holmes: There's a standard I EEC 61,850 dash three, which is the electrical and electronic requirements for the deployment of technology inside substations, which defines a set of requirements that you need to meet in order to be able to deploy technology inside a substation. So we work really hard and we partner with Intel and we partner with VMware around building.

[00:41:50] David Holmes: This infrastructure ecosystem that can be deployed in the locations that energy companies need to rent. Sometimes that's marine regulations. If we're putting things on an oil rig or a [00:42:00] tanker or something like that, sometimes it's specific electrical requirements for substations. Sometimes it is security requirements in North America.

[00:42:08] David Holmes: It's nerc sipps In Europe, it's the Anisa standards where we have to be able to explain to our customers how we're going to integrate our technology into what is their critical infrastructure environment. So what do we sell to customers? A range of solutions around substation modernization and digitalization grid modernization.

[00:42:30] David Holmes: Things like advanced distribution management systems, which manage distributed energy resources. Uh, solar and wind, as well as conventional dispatchable generating resources as well. We build systems that help do reservoir characterization. We build high performance computing systems that help model new types of energy sources.

[00:42:51] David Holmes: You know, if you think across the standard set of new technology workloads, whether it's from AI and digital twin model. Grid operations, looking at how you optimize the operation of wind turbines, and sometimes we get involved in really safety critical applications as well. If you think about some of the protection [00:43:10] relays inside substations, the maximum latency for some of those applications is about 40 microseconds.

[00:43:17] David Holmes: So if something bad happens 

[00:43:18] Matt Trifiro: that that's the definition of something 

[00:43:20] David Holmes: that needs edge computing, well yeah. You know, like if something bad happens, then you need your control system to shut it down in 40 microseconds or. So the whole something bad's happened. Let me send a packet off to the cloud. At that point, a three ton transformer has flown three city blocks, and so that's a classic example of where latency demands mean that you have to use edge computing.

[00:43:42] David Holmes: The really exciting part of it though, Is that traditionally I mentioned about the single function devices and how you have a whole bunch of these inside, inside a substation. But moving forwards, what everybody wants to do is move to a virtualized, containerized environment. They want to virtualize all of those single function devices.

[00:43:59] David Holmes: Should 

[00:43:59] Matt Trifiro: all be multi-tenant and running in a cluster. 

[00:44:02] David Holmes: Put it on a small two or three node, highly resilient cluster, and then all of these new workloads, all of these automation, workloads, they're all containerized. So we need an. That's robust, that supports virtualization, that supports containerization, uh, and that will [00:44:20] dramatically reduce the operations and maintenance costs associated with operating substations.

[00:44:27] Matt Trifiro: You know, we, we talked about it and ot, do you think that distinction is artificial and it eventually goes 

[00:44:31] David Holmes: away? If you ask an OT engineer,

[00:44:37] Matt Trifiro: yeah. But they're all, they're all eventually gonna retire. Right? 

[00:44:41] David Holmes: the same thing, or not the so, so they're not the same thing in a way to choose an IT metaphor, you know, like for the longest time we had storage teams and server teams and database teams and networking teams and virtualization teams, and that was their thing.

[00:44:56] David Holmes: Like, I'm the networking guy, I take care of the network. I work with the, I work with the server guys and I work with the storage guys to make sure everything works together. Uh, but that, but that's my thing. I think operational technology is gonna be a discipline that is integrated seamlessly into it. So, whereas now, okay, that's fair.

[00:45:16] David Holmes: In many organizations you'll see that there's an OT group that begrudgingly works with some of the IT guys when they need an ethernet connection to get some data out to a data center. What we'll see is, is [00:45:30] that it, it will become a much more seamless operation and there is always gonna be a role for operational technologists to fundamentally understand the whole process control landscape of how they're managing and operating systems.

[00:45:44] David Holmes: And they will be dependent upon some IT specialists who are able to. The contemporary edge platforms that integrate with the core and onto the public cloud and help manage the lifecycle of, of data throughout that, that ecosystem. It will remain a discipline, but it will be much more closely to aligned with it.

[00:46:03] David Holmes: It, it's 

[00:46:04] Matt Trifiro: interesting how my thinking, you know, I've been in this the edge world for. Well, the infrastructure world for only about six years and, and my evolution of my understanding has really taken some profound leaps. And you know, one of the things that I see is you've got this very simple difference between like the world of.

[00:46:22] Matt Trifiro: Atoms in the worlds of bits. Right. But to some extent, that's actually a real thing. There's like physical objects in the world that occupy space and physically move, whether it's a car or an oil platform or a wind turbine. And there are things that's just, they're just visual representations. The digital twin or a packet on of ethernet or whatever.

[00:46:39] Matt Trifiro: As [00:46:40] we move into this next decade, there seems to be a profound convergence of these things, and we already see it. I mean, to the extent that my TV has a computer in it, yeah. I mean, most things do. We just don't know about it. Right? Or they're not being used like the the meter that's got the commod or pet.

[00:46:54] Matt Trifiro: Capability on it probably has enough power if it had the right software to be, to be measuring at 35 megahertz or whatever you said as opposed to like every 15, every 15 minutes. Like you wrote Star Wars on all Ave. Yes. Right. I mean like we, we should be able do it better than that. The term of art, that's almost a little cringeworthy, but I also like it is cyber physical and you see it a lot in the academic papers, but this idea that that more of our physical world is be kind of going to become cyber.

[00:47:23] Matt Trifiro: I used to ask in the early days, what's your definition of edge computing? And now I don't ask anymore cuz it's, it's, you know, it's a silly question. It is where the digital meets the physical, it's that actual location. And I'm just wondering if you could, you could sort of react to that in, in your world.

[00:47:36] Matt Trifiro: Do you see that? Yeah. I mean, is that a compelling story or am I way off? No, 

[00:47:40] David Holmes: I, I, I kind of like that, that definition, although I, I gotta say for me, edge goes kind of alike a little bit further away as well. I was, I was just reflecting what you said [00:47:50] about computers inside TVs, because I'm, I moved to Texas from the UK five and a half years ago, and a couple of really strange things happened when I moved here.

[00:47:59] David Holmes: I don't have a landline in my house, haven't done for five and a half years. That's unimaginable to me for my whole life. I moved in here, there was a satellite dish on the roof, but I was. Why would I sign up for a satellite service when I've got a gigabit internet connection and I can do everything online and previously, like you remember H D M I, we'd plug in our D V D player, our satellite box, our Apple tv, and now like you've got this computer inside your TV that does everything for you.

[00:48:27] David Holmes: That era where we had like seven different H DM I cables coming out the back of our tv, those things are long gone and I think. It's really changing the way in which we think about that boundary between the, the physical and the digital. So, and, and in an operational technology world, of course they have kind of, they have sensors, they have valves, they have controls.

[00:48:52] David Holmes: Those things will, will connect to some kind of control device, which will connect to some kind of gateway. And the funny thing [00:49:00] is, Is that on the, on the far left hand side where you've got a little micro controller that's turning a valve you can go, that's very edgy. That's right on the edge. , right?

[00:49:12] David Holmes: That's a definition 

[00:49:12] Matt Trifiro: of edge computing, 

[00:49:13] David Holmes: right? Your point. It's like that's the, that is the literal intersection between physical and digital. There's a microcontroller that's physically turning a valve, and then you get to the gateway, which is consolidating and aggregating all of this information. You're like, well, that's definitely still pretty edgy because that's connecting to the things that's connecting to the physical things.

[00:49:33] David Holmes: Okay, so that gateway then controls to a field appliance. Out in the field, which is, you know, in the middle of nowhere, it's a couple of computers in a shed. And you're like, well, that still feels pretty edgy to me as well. Okay. Right. So then that goes to a regional data center, right. At 

[00:49:54] Matt Trifiro: which point is it not the edge anymore?

[00:49:55] Matt Trifiro: And it's not, it's a infinite 

[00:49:57] David Holmes: continuum. Yeah. So it, it's absolutely continuing. Now, I, I work with a big energy company in Europe. Their definition is really, , the edge is not the public cloud. 

[00:50:07] Matt Trifiro: Yeah. I think that's a bad definition. That's called on-premises. We've been doing [00:50:10] that 

for 

[00:50:10] David Holmes: three decades. Well, not necessarily right, but, but, but the thing is, for them that, that's just how they've chosen to define it.

[00:50:15] David Holmes: There is 

[00:50:16] Matt Trifiro: a, a usefulness between that distinction, but I think all that goes away. That's my personal opinion. I don't know when, maybe not in my professional lifetime, but we're gonna stop talking about the edge. Except as a, you know, in a very, very small group of people, it's just like, is it on 

[00:50:30] David Holmes: the network or not?

[00:50:31] David Holmes: Yeah. And, and so, and so where that gets you to is think about the conversations we've been having about hybrid cloud, about multi-cloud, super Cloud. People are starting to talk about now Cloud Connect, near cloud co-location providers. You know, you can slice and dice this. One of a number of different ways, but it's no longer a useful.

[00:50:53] David Holmes: There is gonna be compute and data across the entire continuum, right from that little valve all the way up to a hyperscaler data center. And the people who are gonna be successful are the people who work out where the right technical and economic places to run all of these different workloads. 

[00:51:11] Matt Trifiro: You hit it, you 

[00:51:13] David Holmes: hit the nail on the head because.

[00:51:14] David Holmes: That's where we're at, you know, like five, six years ago. It was almost like a kind of it, it was almost like a [00:51:20] religious war. It's like I believe in one cloud. My chosen provider is AWS Cloud. How many press release hybrid Cloud, right? How many press releases did you see where XYZ company enters into partnership with XYZ public cloud provider to transform their digital assets to shut down their data centers delivering unimaginable.

[00:51:43] David Holmes: Productivity enhancements while massively reducing costs. Very excited. This five year deal valued at over 500 million, blah, blah, blah, blah, blah. All of the major energy companies you can, and they're all written in the same form. And what they discovered is that some of this stuff is hard. Some of this stuff does not make sense.

[00:52:01] David Holmes: Now the debate about the role of the public cloud in utility computing is, is long gone. It has an incredibly important. But as we look at what the, where the future of computing's going, We're gonna see far more computational capability, the edge of the networks than we are inside the hyperscaler data centers.

[00:52:22] David Holmes: And much more data's gonna be generated there as well. 

[00:52:25] Matt Trifiro: Yeah. And the reality is that might be running on a Dell server, it might be running on an Amazon [00:52:30] server, it might be running on an Amazon server, but from Dell. Right? I mean, it's a really interesting world. David. We have gone over our hour. We could talk.

[00:52:38] Matt Trifiro: Three hours. I think at some point I'll have to have you on the show to, to see how the world's changed, but it's exciting to see the work that you're doing personally, that your division's doing. And it's interesting to learn more about Dell cuz I, I do think of Dell back in the customizable desktop that's in the fold out in the back of the magazine.

[00:52:56] Matt Trifiro: And it's a, it's a much more divers. Company now as illustrated by this entire energy 

[00:53:02] David Holmes: industry that you're running? Yeah, I mean, Dell is now a hundred billion dollar, one of the, the largest technology providers, the fully integrated portfolio from client storage, data protection, server compute, multi-cloud.

[00:53:16] David Holmes: It's a compelling powerhouse of solutions. And the thing that gets me up every morning is this, I've got three goals, number. Accelerate the energy transition. Number two, advanced decarbonization and number three, ensure global energy security. And I have the entire resources of Dell's engineering portfolio to help me build the solutions that are gonna let me do those three things.[00:53:40]

[00:53:40] David Holmes: And that's why I think I have one of the coolest jobs that's available in technology today. And I hope other people think exactly the same thing about their job. 

[00:53:48] Matt Trifiro: Well, the way, the way you just put it, that's, that's pretty infectious. But they're, I'm right cuz 

[00:53:51] David Holmes: I got the coolest 

[00:53:52] Matt Trifiro: job. Well then I have the second coolest job.

[00:53:53] Matt Trifiro: I'll go with that. All right, David, thank you so much for being on the podcast. 

[00:53:57] David Holmes: Thank you so much. It's. 

[00:53:59] Narrator 2: That does it for this episode of Over the Edge. If you're enjoying the show, please leave a rating and a review and tell a friend. Over the Edge is made possible through the Jenna sponsorship of our partners at Dell Technologies.

[00:54:11] Narrator 2: Simplify your edge so you can generate more value. Learn more by visiting dell.com.