In our third episode, David Sloan, Chief Technology Officer for Microsoft Worldwide Public Sector, joins Christopher McKenna and Natasha Allen for an in-depth discussion about hyperscale cloud computing. What is it and how did it power the rapid digital transformation during the COVID-19 pandemic? How does it differ from “standard” cloud computing and what benefits does it offer by comparison? Where does it intersect with areas of investment such as artificial intelligence and cybersecurity?
- Competing Interests: Data Ownership at the Crossroads of SaaS Stakeholders and Regulation
- Key Strategies For Patenting Big Data Solutions
The below episode transcript has been edited for clarity.
Hi everyone and welcome. My name is Natasha Allen and I am a partner in Foley’s Silicon Valley office and the Co-Chair of the AI area of focus within our Innovative Technology sector. On today’s podcast, we’ll be discussing the hyperscale cloud and associated technologies. Joining me to provide their expertise on the topic are Dave Sloan and Chris McKenna. Good morning gentlemen.
Good Morning Natasha.
Dave Sloan is Microsoft CTO for the Worldwide Public Sector focused on Global Market Development. In this role, he works with governments around the world to create architectures for innovation that are compatible with national priorities and regulations. His particular areas of focus are tailoring cloud adoption to comply with national priorities, flexible cloud architectures, advanced data protection measures, and hybrid innovation.
Chris McKenna is Foley’s Office Managing Partner in Boston, Co-Chair of the Electronic Practice, and Chair of the Cloud Computing Infrastructure and Solutions within our Innovative Technology sector. Chris focuses his practice on acting as advisor for technology companies, helping identify, capture and protect their most important IP. In addition, Chris counsels companies on legal, technical and business issues of IP, in various transaction documents and various transactions, such as M&A deals.
So first question, what is the hyperscale cloud?
Let’s start with the fundamentals. People may say that the cloud is basically utilizing somebody else’s servers. Instead of running your own infrastructure for your own enterprise, the task is outsourced to another organization. That is one aspect of it, but the hyperscale cloud has capacity far beyond that, creating enormous economies of scale. Hyperscale appears to offer near limitless capacity for customers that they can access on demand, pay as you go, with the latest and greatest technology kept evergreen without any of their intervention. That is really the definition of the hyperscale cloud – which is tantalizing for both private and public sector customers as they look to found their IT plans for decades to come.
I’d like to add a couple of comments. There’s somewhat of a misconception of what the cloud is and the first place. It’s just not the location of data and applications, from on-premise, in your data center, to a third-party online resource. It’s also a style of computing. It’s scalable and elastic, and can provide resilience and many capabilities as a service to customers using internet based technologies. To Dave’s point it’s an architecture that scales according to demand.
I do have a question for Dave. Every time I hear hyperscale, I always associate it with, or I hear people associate it with the public cloud. Would hyperscale include private and hybrid cloud models as well – as long as they’re scalable or does the industry really see it as the public cloud?
No, I think the transformational economics of hyperscale IT really only apply to the hyperscale cloud. They certainly don’t apply private clouds. And they apply to hybrid cloud, I would say, in very limited ways. It would suffice to say that the goal of hybrid cloud is to bring as many aspects of those hyperscale benefits into a private cloud scenario as possible – but the upper limits of those possibilities are still extraordinarily constrained.
Thanks Dave. That makes a lot of sense.
It seems like there’s a pit against hyperscale cloud providers and IT service providers. How is hyperscale cloud technology adoption changing the IT field?
It’s become the new normal. This is a revolution in computing. It’s really a remarkable generational shift, although we’re still only at the leading edge of it. I would identify three key aspects that have had a transformational effect on IT. Happy to go into more detail on any of them, if it’s of interest. They are agility, innovation and cybersecurity.
Cybersecurity is sort of the negative driver. You know the threat environment currently is transforming at a pace and scale that you’ve never seen before. The only safe refuge is those who can fight it at scale, by taking comfort in the protection that the hyperscale cloud allows.
Agility, the ability to fail fast and cheap, to experiment with IT in ways that may have previously required many, many months, and in some cases, many millions of dollars, was historically a barrier to entry. That has been completely shattered. You can have an idea on a Tuesday and have it stood up on a Thursday. If it doesn’t work, then shut it down on a Monday – having expended minimal time and money to iterate, or even to launch simultaneously. Agility is key these days, given the unforeseeable and unforeseen demands that our public sector customers are dealing with. Pandemics, war, shifting alliances, you name it – agility is our brand.
The third aspect, innovation, is just unlocking the previous two. We have a breadth, depth, and ease of adoption of an unbelievable array of innovative technologies that can go to market faster and more broadly, using the hyperscale cloud, than before. Some of those innovative techniques will only be available on the hyperscale cloud because it’s no longer marginally profitable for innovators to offer them in other contexts.
Those are sort of three transformative aspects that I would hit on: cybersecurity, agility and innovation. Chris, I don’t know if you have others.
You said it well. One thing I think of is real world constraints. It’s always people and money. Cloud computing represents a shift from the expensive physical IT resources of on premise. That requires a certain amount of people and equipment that’s not really scalable or tied to scale to a point – unlike being able to stand up using SCRM resources. That means solutions in hours and days versus weeks and months like the typical enterprise models. The shift to cloud computing gives a much smoother, secure, controllable online resource that can benefit way more people.
There’s been a big shift just in terms of scalability. To your point, the pandemic proved that – in terms of getting some of our basic physiological needs of safety, food, and connecting people together. As we come out of this pandemic, we will continue to see high-level innovation due to the leverage the cloud creates and the ease of adoption. Consumers got a taste of it being at home. Even the folks who may not have been early adopters to technology find themselves using mobile apps and online services. Older generations and folks of all different demographics are now all on the cloud. It’s become an expectation of our citizens to have services that are available 24/7.
That leads us into our next question. Based on what you said, Chris, do you think that hyperscale data centers and operations will continue and persist past the pandemic?
The pandemic really tested the world’s infrastructure. We’ve seen traditional companies, who did not have a cloud-first strategy, have to get online to survive and keep themselves relevant during the pandemic. Going back to my previous point, we saw consumer adoption of cloud and on-demand services increase. People using Zoom and Amazon for example – getting all your needs just shipped to your house. As we come out of the pandemic, I see continual digital transformation. We have better infrastructure now because we had to adapt to the online surge. We have more companies who are going to have a cloud-first strategy.
Instead of innovation being focused on the pandemic alone, I believe there will be further evolution. Think of Maslow’s Motivation Model. During the pandemic, we were taking care of physiological needs, safety needs, belonging and connection needs, the bottom of that pyramid. Now we’re going to go up the motivation model to other things – for quality of life and convenience.
I’ll pick a couple areas: health and services. First, we had telemedicine providing basic health care at home, with the ability to attend basic doctor visits to diagnostic tests. That accelerated to the point where those features are now expected. Thanks to the cloud, we now have home health devices not just for safety, but general health. You can do blood work at home, test glucose, and even treat a chronic illness.
Anything that’s a service is going to grow across sectors – from retail to transportation. In this decade, we’ll get to the point of having autonomous vehicles coming to pick people up, moving goods and products around – all with a simple click. We’re at the tip of the iceberg in terms of the level of innovation we’re going to see coming out of the pandemic.
Yeah, I mean, this car has one gear and it’s drive. You can decide how fast you want to go, but it’s just going forward. Our president of Microsoft, Satya Nadella, characterized the beginning of the pandemic as having seen three years of digital transformation occur in three months.
That pace has basically kept up. Obviously, I focus very much on the public sector where the demands are growing, both in variety and severity at a breakneck pace. Budgets are more constrained than they’ve ever been before. The cloud unlocks so many possibilities for public servants to be able to do more with less – delivering more and higher quality public services. Even in an era of constrained or shrinking budgets, there really is no alternative to the cloud.
All of the innovations that have been brought up came into play during the pandemic and need the support of hyperscale data center cloud computing. With new innovation comes legal issues. How do you navigate? How do you handle this compliance environment and the innovations that come with? What are some of the headaches that come with cloud computing on hyperscale?
One word we’ll talk about is governance. From a legal compliance perspective, we have the contractual side of the equation, saying what you’re going to do, then all these vendors’ sign up for contractual terms. It could be privacy, security, IP, data management, financial management, or performance management. There’s a whole bunch of different aspects of that. You sign up contractually. On the other side of the equation, you have operations. You need to do what you said you were going to do and be compliant.
Cloud governance is an important part of that process. It’s a set of rules and policies that companies use to manage their services. And it has different aspects. There’s financial management, operation management, security and compliance management, data management, performance management, and asset and configuration management.
When I think of quality management and cloud governance, you have processes, people, and technologies. You need to have those policies in place. You have to train people on those policies and you have tools to manage. Vendors like Microsoft have tools that monitor compliance across multiple different aspects. You need people trained on those tools to do that.
The holy grail is to be 100% compliant, 100% on time. I kind of equate it to managing bugs in software. That’s really hard to do, but the goal with cloud governance is to do that.
And there’s actually an emerging software domain in cloud governance software management tools because of the many complexities that come with it. You also need a lot of talent to achieve your goals. I think the talent pool needs to change for the cloud to address more enterprise needs.
First of all, it’s important to make some category distinctions here between the consumer and enterprise portions and usage of the hyperscale cloud. There’s more change to come in the exploitative models of the cloud, rather than the enabling models of the cloud. It’s really too easy to throw out the baby with the bathwater in these regards, if we are not willing to apply a little bit of scrutiny and a little bit of distinction in this space.
As the world trends towards valuing privacy as a fundamental right and ensconcing that value in legislation and regulation, we need to understand that our technological infrastructures and architectures need to evolve to accommodate that fundamental right.
From an international relations perspective, as we establish the United States as military superpower and China as a financial superpower, Europe has established itself as a regulatory superpower in and we certainly see echoes and emanations of the European regulatory state, specifically on cloud regulation. And Europe’s not done. We need to be careful, in this regard, to make sure that we are understanding that, while privacy is a fundamental right, it is one of many fundamental rights. That balanced compromise is inevitable.
In the meantime, there’s also very practical conversation. As more twigs are added to this compliance nest, it becomes more and more burdensome for companies and agencies and ministries and departments that want to innovate, that want to digitally transform, to even demonstrate, that they are doing so in a compliant fashion.
It is becoming a full employment act for lawyers, but becoming unreasonable burden – especially to small and medium sized businesses. To be even those that have no intention of doing anything exploitative or anything transgressive, just demonstrating positive intentions – has become an extraordinarily slow, time consuming and expensive process.
From my perspective, the hyperscale cloud has sort of a democratizing effect. Once you ally yourself with one of the big providers, you also ride their coattails to a certain extent. Enterprises can say “if the platform that I am computing on has demonstrated its compliance according to regimes worldwide, I just have to establish the compliance of whatever additional work I’m doing on the cloud” and that’s a much more feasible task for an entrepreneur or a small-medium sized business to be able to achieve. So, I think there’s some synergy here, between, expanding compliance regimes and expanding hyperscale cloud adoption. At the same time, it is a very dynamic space and we are looking forward to very turbulent times worldwide.
On a high level, I saw an article announcing that Microsoft execs were reaching out to Congress for a federal level privacy framework to follow. In other countries, the regulatory schemes are ahead of us and the U.S. needs to play some catch up. Wherever you are on that supply chain, you have multi-jurisdictional issues of compliance to deal with.
Let’s just focus on the United States and just about privacy. Right now we have a patchwork of privacy laws on a state basis. I could be a U.S.-only cloud provider leveraging services like Microsoft and all of a sudden you have a patchwork of state level privacy laws that you have to deal with in terms of delivering cloud services. I think the government can do something to help that along, in terms of regulatory regimes and making it easy, while being respectful to privacy. At the same time, we need to make the adoption of cloud services accessible for everybody.
The innovation supply chain often starts with the small to medium-sized players – innovators who can’t afford costly regulatory hurdles to accessing the cloud. As Dave mentioned, aligning yourself well with a cloud provider, who has all that compliance capability built in, and you just have to do the delta. It gets you pretty far.
It’s a complex political environment and there’s a lot of countervailing winds. But at the end of the day the business of America is business. And this chaotic patchwork of different standards across America is bad for business, right? Harmonization benefits pretty much everyone. I think the sooner that the U.S. is able to come to a consensus on what’s appropriate, at least as a minimum standard, that’s going to be good news for business and the public sector.
It’ll have tremendous benefits for multi-national corporations as well. We won’t get into some of the adequacy determinations with the EU and some of the more in the weeds issues today.
Thank you. It is a very interesting landscape. Here’s my final question. We know that AI is permeating pretty much every industry and every sector. How does the use of AI play out in the cloud technology environment?
This goes back to the point I mentioned before, about where innovation is happening. AI today is deeply entwined with big data. It requires the capacity to be able to analyze enormous amounts of signal in order to be able to come to its magical conclusions. In many cases it requires specialized hardware and esoteric architectures. It then becomes prohibitive, for any individual enterprise, public or private sector, to be able to invest in, maintain, and update those architectures, to be able to pursue and deploy modern artificial intelligence techniques.
As a result, you see a concentration of AI activities on the hyperscale cloud, where the economies of scale work and where investment in AI actually provides a competitive advantage. It becomes a virtuous cycle. The flywheel can attract even more and more capability – which is just good news for customers, whether they’re in the public or private sector. It supercharges not only innovation, but the accessibility of that innovation to a broad swath of stakeholders in very rapid forms. AI is already concentrated tremendously on the hyperscale cloud. That’s a trend I see increasing, not decreasing, over time.
To add my flavor to the topic, one purpose of the cloud is to make services ubiquitous. And to scale you need a large audience base. One way to customize that digital experience for these users of different demographics is through AI. AI can learn about users in a systematic and automated way to improve a customized set experience. With digital transformation at the hyperscale, massive data can get ingested, processed, transformed, stored for these AI models.
When you put AI and big data and hyperscale cloud together, its business intelligence on steroids. Business intelligence services are used in cloud based AI services to get those deep insights into the behavior of their target audience to make the whole overall user experience better. Innovations are also going to be driven from that AI experience.
Data is the new form of innovation IT currency. Having a data governance model that enables one to monetize that data, in a way that’s respectful to privacy, but still allows the service provider to improve their products and services, is really valuable.
That’s great. Well, thank you both. Are there any departing words that you have for our listeners before we wrap up?
Watch this space. There’s a tremendous amount of change coming in the short term. There’s also uncertainty about where these interplays of regulatory and technological innovation are going to pan out in the long term. Finally, there are lots of private and public sector customers who desperately need good counsel and advice.
It’s an exciting time. I think, as a human race, we’re going to see so much innovation in this decade. Ten years from now, we’re going to look back, and see how much we’ve grown. We’re at a transformational era that’s only going to improve our quality of life. And as we discussed, companies navigating that process will need strong business and legal counsel to understand the multi-jurisdictional landscape.
Thank you both. I really appreciate you taking the time to talk with me. It was a very informative talk.
Foley & Lardner’s Innovative Technology Insights podcast focuses on the wide-ranging innovations shaping today’s business, regulatory, and scientific landscape. With guest speakers who work in a diverse set of fields, from artificial intelligence to genomics, our discussions examine not only on the legal implications of these changes but also on the impact they will have on our daily lives.