(upbeat music) >> Sustainability has become one of the hottest topics, not just in enterprise tech, but across all industries. The relentless pace of technology improvement over the decades and orders of magnitude increases in density have created heat, power and cooling problems that are increasingly challenging to remediate. Intense efforts have been implemented over the years around data center design techniques to dissipate heat, use ambient air, liquid cooling and many other approaches that have been brought to bear to get power usage effectiveness, PUE, as close to one as possible. Welcome to Better Together Sustainability, presented by the CUBE and brought to you by Hewlett Packard Enterprise and AMD. In this program we'll lay out today's challenges and how leading companies are engineering solutions to the problems just introduced, along with some recommendations, best practices and resources as to how you can initiate or enhance your sustainability journey. First up to help us better understand this important topic are John Fry, senior technologist IT efficiency and sustainability at Hewlett Packard Enterprise and Terry Richardson, North America channel chief for AMD. Gents, welcome. >> Great to be here. >> (indistinct). >> John, let's start at the high level here. Why is sustainability such an important topic today? Why now? Why is it such a challenge for customers and, and how are you guys approaching the solutions? >> The topic has been an important topic for a number of years, but what we're seeing across the world is more and more corporations are putting in place climate targets and sustainability goals. And at the same time, boards and CEOs are starting to be asked about the topic as well, making this topic much more important for technology leaders across the globe. At the same time, technology leaders are fighting with space, power and cooling constraints that caused them to rethink their approach to IT. To get a better sense of how wide this challenge is, we did a survey last year and we asked 500 technology leaders across the globe if they were implementing sustainable IT goals and metrics and programs within their infrastructure. Personally, I thought the answer would be about 40% of them had these programs. Actually it turned out to be 96% of them. And so when we asked them why they were implementing these programs and what was the primary driver, what we heard from them was three things. Those of them that were the early adopters and the ones that move were moving the fastest told us they were putting these programs in place to attract and retain institutional investors. If they're a publicly traded company, their investors were already asking their boards, their CEOs, wanting to know what their company was doing to drive efficiency within their technology operations. Those companies in the middle, the ones that were just moving along at the same pace as many other companies around the world, told us they were putting these programs in place to attract and retain their customers. Customers are increasingly asking the companies they do business with about their sustainability aspirations specifically how technology contributes to their carbon emissions and their sustainability goals. And so these customers want to make sure that they can keep their own customers. And finally, a third group, the digital followers, that group of companies that's a little slower adopting programs, more conservative in nature. They said they were implementing these programs to attract and retain employees. In fact, over the last year or so, every customer we've talked to when they describe their pain points and their challenges that we can try to help them meet, has had a difficulty in finding employees. And so what we know is these younger employees coming into the workforce, if you can show them how what they are going to be doing connects to the purpose of their company and connects to making the world a better place, you can attract them easier and you can retain them longer. So a variety of business reasons why companies are looking at these programs, but what we know is when they implement these programs they often reduce over-provisioning. They save money, they have a lower environmental footprint, and again they have an easier time attracting and retaining employees. So for all of these reasons, driving sustainability into your IT operations is a great thing to do. >> Yeah, I never would've expected 96%. And of course, investors, customers and employees. I mean, this is the big three. Terry what's AMD's perspective on this topic? In other words, what do you bring to the table and the partnership? I mean, I know processors, but what's unique about AMD's contribution? >> Yeah. Thanks Dave. And, and John, great to be with you. Appreciate the opportunity and the partnership. You know, we too are very focused on sustainability and enjoy our partnership with HPE very much in this area. You know, since 2017, when AMD introduced its epic processor family, there's been a couple of core design elements in that technology. One has to do with performance. And the second has to do with efficiency. Both are critically important to today's topic of sustainability because increasingly, customers are understanding and measuring performance per watt and fortunately, AMD really excels in this area. So whether we're talking about the larger super computers in the world, or even general purpose servers, customers can fundamentally do more with fewer AMD servers than competitive alternatives. And so, so we, we really bring a technology element on the processor side, CPU and GPU, to play a role in delivering real ability for customers to meet some of their core sustainability goals. And of course, in partnership with HPE, together we have really a compelling story. >> Great. Thank you, Terry. And, and John, wonder if you could talk to the differentiation that you bring from HPE's perspective, the total package. >> Yeah, of course. The first thing as partnership. As Terry mentioned, AMD and HPE have been working together since HPE was founded actually, to drive power efficiency up to meet the demands of our customers. At the same time, as our customers have asked more and more questions around technology sustainability, we've realized that we needed to not only develop a point of view on that from an HPE perspective, but actually write the white papers that give the customer guidance for sustainable IT strategies, for tech refresh cycles, give them some guidance on what are the right questions to ask technology vendors when they're buying technology equipment. So a series of white papers and you might not appreciate why, but this is a topic that you can't go get a college degree in and frankly can't even buy a book on. So for customers to get that knowledge, they want to get it from experienced professionals around the globe. And in fact, in the survey that I mentioned earlier, we asked customers, where's the number one place that you expect to get your sustainable IT information from? And they said, our technology vendors. So for us, it's really about driving that point of view, sharing it with customers, helping customers get better and even pointing out some of the unintended consequences. So a great example, Dave, you mentioned PUE earlier. Many customers have been driving PUE down for a number of years, but often the way that they did that was optimizing the data center building infrastructure. They got PUE pretty low. Now, one of the things that happens and customers need to be aware of this, particularly if they're focused on PUE as their primary metric, is when they optimize their IT stack and make that smaller, PUE actually goes up. And at first they think, well, wait a minute, that metric is going in the wrong direction. But when you remember it's a ratio, if you get that IT stack component smaller, then you're driving efficiency even if PUE goes the wrong direction. So part of the conversation then is you might want to look at PUE internally, but perhaps you've outgrown PUE and now have an opportunity to look at other metrics like carbon emissions per workload, or or power consumption per piece of equipment or rack. So all of this drives back to that upward trajectory that Terry was talking about where customers are really interested in power performance. So as we share those stories with customers, share the expertise how to move along this journey, that really provides great differentiation for HPE and AMD together. >> So that's interesting. So PUE is not necessarily the holy grail metric. There are other metrics that you, you should look at. Number one, and number two the way you interpret PUE is changing for the better. So thank you for that context. I wonder Terry, do, do you have any like proof points or examples that you can share? >> Yeah, so one that immediately comes to mind that was a manifestation of some terrific collaboration between AMD and HPE was their recently announced implementation of the Frontier supercomputer. That was a project that we collaborated on for a long time. And, and where we ended up was turning over to the government a supercomputer that is currently the highest performing in the world, broke the exaFLOP barrier. And probably even more importantly is number one on the Green500 list of the top super computers. And, and together we enjoy favorable rankings in other systems, but that's the one that, that really stands out in terms of at scale implementation to shine really a spotlight on what we can do together. Certainly for other customers doesn't have to be the world's largest super computer. It's not uncommon that we see customers just kind of in general purpose business applications in their data centers to be able to do more with less, you know, meaning, you know, you know a third of the servers oftentimes delivering not only a very strong TCO but the environmental benefit that gets associated with significantly reduced energy that can be expressed in reduction in, in overall CO2 emissions and other, other ways to express the benefit, whether it's, you know the equivalent of, of planting you know, acres of forest or whatever. So we're really proud of the proof points that we have and and look forward to the opportunity to together explain this more fully to customers and partners. >> Right? So John, Terry sort of alluded to this being more broad based. I know HPE has a very strong focus on HPC. Sorry for all the acronyms, but high performance computing. But the, so this is more broad based than just the super computing business, right John? >> Yeah, absolutely. We see these performance benefits for customers and industry standard servers as well. In fact, many customers, that's the primary type of equipment they use and they want better power performance. They either want to as Terry alluded to, use less equipment to do the same amount of work, or if they've run into a space or power or cooling constraint in their data centers, they want to be able to increase workloads in the same footprint. So it allows them to take better use of their data centers. And for some customers even the data center enclosure that they started with they can actually use a much smaller amount of space. In fact, we have some that even move over to co-location facilities as they improve that performance per watt, and can do more work in the smaller space. So it starts an industry standard server, but increasingly we're seeing customers considering liquid cooling solutions and that generally moves them into the high performance compute space as well right now. So those performance improvements exist across that entire spectrum. >> So since you brought up liquid cooling John, I mean can you share any best practices? I mean, like what do you do with all that heated liquid? >> Yeah, it's a great, great question. And we have seen a lot more interest from customers in liquid cooling and there's a variety of things that you can do, but if you're considering liquid cooling the opportunities to think broader than just the IT stack. So if you're going to use a cooling loop anyhow and you're going to generate warm liquid coming off the it equipment as waste, think about what you can do with that. We have a, a government customer here in the United States that designed their high performance computer while they were designing the building it went in. So they're able to use that hot air, hot water, excuse me coming off the IT equipment to heat the entire building. And that provides a great use of that warm water. In many parts of the world, that warm water can either be used on a hot water utility grid or it can even be used on a steam grid if you can get it warm enough. Other places we're aware of customers (indistinct) and greenhouses next to data centers and using both the warm air and the warm water from the data center to heat the greenhouse as well. So we're encouraging customers to take a step back, look at the entire system, look at anything coming out of that system that once was waste and start to think about how can we use that what was waste now as an input to another process. >> Right, that's system thinking and some, some pragmatic examples there. Can, can you each summarize, maybe start Terry, with you AMD's and HPE's respective climate goals that may, Terry then John chime in please. >> Yeah, I'll go first. We actually have four publicly stated goals. The first one is I think very aggressive but we've got a track record of doing something similar in our client business. And, and so kind of goal number one is a 30 X increase in energy efficiency for AMD processors and accelerators powering servers for AI and HPC by 2025. The second is broad based across the corporation is a 50% absolute reduction in greenhouse gas emissions from AMD operations by 2030. And then the third is 100% of AMD manufacturing suppliers will have published greenhouse gas emissions reduction goals by 2025. And we've declared that 80% or greater of our manufacturing suppliers will source renewable energy by 2025. Those are the, those are the four big publicly stated goals and objectives that we have in this area. >> You know what I like about those Terry? A lot of, a lot of these sustainability goals these moonshot goals is like by 2050, it's like, okay. But I, I like the focus on '25 and then of course there's one in there at the end of the decade. All right, John, maybe you could share with us HPE's approach. >> Yeah, absolutely. And we've had almost two decades of emissions reduction goals and our current goals, which we accelerated by 10 years last year, are to be carbon new or excuse me, net zero by 2040. And that's a science based target-approved goal. In fact, one of the first in the world. And we're doing that because we believe that 2050 is too long to wait. And so how we reach that net zero goal by 2040, is by 2030, an interim step is to reduce our scopes one and two, our direct and energy related emissions by 70% from 2020. And that includes sourcing 100% renewable energy across all of our operations. At the same time, the bigger part of our footprint is in our supply chain and when our customers use our products, so we're going to leverage our as a service strategy HPE GreenLake and our energy efficient portfolio of products to reduce our scope three carbon emissions 42% over that 2020 baseline by 2030, and as with AMD as well, we have a goal to have 80% of our suppliers by spend have their own science based targets so that we know that their commitments are scientifically validated. And then the longer step, how we reach net zero by 2040 is by reducing our entire footprint scopes one, two and three by 90% and then balance the rest. >> Yeah. So again, I mean, you know 2030 is only eight years away, a little more. And so if, if, if you have a, a target of 2030 you have to figure out, okay, how are you going to get there? The, if you say, you know, longer, you know in the century you got this balloon payment, you know that you're thinking about. So, so great job, both, both companies and and really making more specific goals that we can quantify you know, year by year. All right, last question, John. Are there any resources that you can share to help customers, you know, get started maybe if they want to get started on their own sustainability strategy or maybe they're part way through and they just want to see how they're doing. >> Yeah, absolutely much of what Terry and I have talked about are available in an executive workbook that we wrote called "Six Steps For Implementing a Sustainable IT Strategy" and that workbook's freely available online and we'll post the URL so that you can get a copy of it. And we really developed that workbook because what we found is, although we had white papers on a variety of these topics, executives said we really need a little bit more specific steps to work through this and implement that sustainable IT strategy. And the reason for that, by the way is that so many of our customers when they start this sustainable IT journey, they take a a variety of tactical steps, but they don't have an overarching strategy that they're really trying to drive. And often they don't do things like bring all the stakeholders they need together. Often they make improvements without measuring their baseline first. So in this workbook, we lead them step by step how to gather the right resources internally, how to make the progress, talk about the progress in a credible way, and then make decisions on where they go next to drive efficiencies. >> Yeah, really that system thinking is, is, is critical. Guys. Thanks so much for your time. Really appreciate it. >> Thank you. >> Okay guys, thanks for your time today. I really appreciate it. In a moment, We're going to toss it over to Lisa Martin out of our Palo Alto studio and bring in Dave Faffel, chief technology officer at WEI, along with John and Terry, to talk about what WEI is doing in this space to address sustainability challenges. You're watching Better Together Sustainability brought to you by HPE and AMD in collaboration with the CUBE, your leader in enterprise and emerging tech coverage. (lilting music)