The Future of Digital Infrastructure – Interconnections – The Equinix Blog

Digital technology continues to be a critical driver of economic and social progress. It is creating extraordinary business opportunities and helping to elevate the standard of living for billions of people all around the world. Digital is unleashing the limits of human potential and, in the process, stimulating unprecedented demand for a new generation of digital infrastructure to meet the rapidly changing demands of our global society.

When Equinix established our first vendor-neutral, multitenant data center 25 years ago, we pioneered a new approach to digital infrastructure, helping to unlock the power of digital technology in ways that we never could have imagined. Today, we celebrate a quarter century as a cornerstone of digital transformation, proudly supporting our customers in enabling what matters most in a world defined by connectivity.

From the beginning, it was our mission to act as a steward for some of the world’s most important digital infrastructure, creating a place to connect and exchange data securely, allowing digital ecosystems to thrive. To reflect this mission, our founders chose the name Equinix to underscore our commitment to offering an EQUal, Neutral Internet eXchange.

This helped set in motion a fundamental shift in how the world consumes digital technology. In the past, data centers were siloed and centralized, built by and for the benefit of a single organization. Today, those traditional data centers are being replaced by highly interconnected internet exchange (“IX”) points distributed throughout the world. These IX points are now the engine room of the digital economy. They help businesses tap into digital ecosystems to scale their operations, fuel innovation and access new markets quickly. The growth of interconnection bandwidth shows just how essential digital ecosystems have become.

Click to Enlarge

A big milestone like this is an opportunity to reflect on and celebrate what we have accomplished, but also the perfect moment to look ahead at what the next 25 years might bring. To say I’m optimistic about the future would be an understatement. Here are just a few of my expectations for the future:

• Digital has been changing the lives of consumers for decades, and that wave of transformation is now changing the way business is done. Digital transformation is reshaping the basis of competition in nearly every industry vertical, separating those that are prepared for the new digital reality from those that are not and, in the process, shifting the future dynamics of the global economy. Emerging technologies like AI will add fuel to the fire, further accelerating the need for transformation—and stimulating demand for modernized infrastructure to support that transformation.
• Our industry will continue to create the platforms that digital ecosystems need to thrive. Leveraging globally distributed infrastructure and the power of software, we plan to make it faster, easier and more frictionless to assemble and deploy hybrid infrastructure, enabling customers to exchange value and leverage each other’s services in new ways.
• We’ll continue working to advance digital inclusion worldwide—creating new opportunities in underserved communities and unlocking the full potential of our global village. Achieving this objective will require thoughtful support and deep collaboration across businesses and government to help carry the benefits of digital into the places they’re needed the most. We launched the Equinix Foundation to support digital inclusion initiatives throughout the world, and we are excited to see others making similar commitments.

I’ve asked experts from across Equinix to join me in sharing their vision for the future of the digital world. Read on to see their thoughts on the biggest opportunities ahead, the new challenges businesses will need to overcome and how digital infrastructure will evolve to keep up with those changes.

Over the next 25 years, I expect businesses will move away from owning digital infrastructure and toward acquiring it via subscription. An Everything as a Service (“XaaS”) model will become more desirable because infrastructure will advance quicker. Buying your own hardware will become a competitive disadvantage: You’ll either keep using it long after something better becomes available, or you’ll refresh early and risk not getting a return on your initial investment. Meanwhile, businesses subscribing to their infrastructure will be able to refresh and improve continuously.

Deploying and managing hardware won’t be for amateurs anymore. You’ll be better off partnering with specialty service providers that work with that infrastructure daily. They’ll help you get what you need in an automated, programmatic way. This means that deploying infrastructure and setting up new business models will be as quick and easy as consuming software is today. As new technologies emerge, you’ll subscribe to those technologies on demand from any device. The only limit will be your imagination.

This is in stark contrast to today, a time when many organizations fear the cost and complexity of adopting new technologies such as AI. The Equinix 2023 Global Tech Trends Survey (“GTTS”), a survey of 2,900 IT decision makers polled worldwide, found that 46% of global IT leaders said that increased OPEX would deter them from adopting a newer technology, while 37% said the same regarding slow implementation. These and other deterrents leave organizations unprepared to keep up with the accelerated pace of innovation that we’ll see in the years to come. However, the shift to subscription-based digital infrastructure will make many of these concerns a thing of the past.

The digital infrastructure industry provides the platforms on which ecosystem partners gather to exchange data with one another quickly and securely. Twenty-five years from now, I expect all business will be digital business, meaning all business processes will be nothing more than software passing data back and forth over networks. That’s why I think our industry will play an increasingly central role in the global economy.

Honestly, I don’t believe the digital infrastructure industry will change that much over the next 25 years. We’ll continue to grow platforms that make it easier for ecosystem partners to exchange more data with each other faster and in more locations. But that’s just a continuation of what we’re already doing, rather than a fundamental shift.

What I believe is going to change is that the business world as a whole is going to become more like our industry, and catch up to the things we’re already doing. Business in 25 years is going to look identical to what digital infrastructure looks like today, just on a much wider scale.

Participating in digital ecosystems will be essential

In the future, what we now know as the digital economy will simply be known as the economy. The traditional approach that ruled the business world for decades will no longer exist. The same forces that made digital transformation important today will make it essential 25 years from now.

In fact, I expect the term “digital transformation” will be long gone from our vocabulary by that point. It will go without saying that all successful businesses will be fully digital. Even today, we’re seeing once-great brands fall by the wayside because they failed to invest in digital, or they couldn’t make the cultural shift needed to capitalize on investments they did make. I expect we’ll see others join them in the years to come.

I think the most successful businesses will be the ones that get the most value from their digital ecosystems. We already say that “your ecosystem is now your infrastructure,” and this trend will only continue to accelerate going into the future. We’ll reach a point where digital ecosystems will be an essential part of every business’ operations.

If you aren’t participating in digital ecosystems by then, you won’t just be at a disadvantage; you won’t even be in the game at all. Anytime a company needs to buy a product or service, it will choose the supplier from among its ecosystem partners. If you’re not in the ecosystem, the business opportunity will be gone before you even knew it existed.

In 25 years, I believe the presence of AI in our everyday lives will be as common and seamless as electricity is today. Every aspect of our lives will be guided by expert systems that exist for the sole purpose of helping humans. Examples of AI systems that I expect will become commonplace include:

• VR-based systems: Humans will use AI-driven virtual reality systems to help do everything from predictive maintenance on vehicle fleets to choosing the right clothes.
• Robotics in everyday life: AI-driven robots will increasingly perform daily chores, including cooking and cleaning.
• Brain/computer interaction: Humans will be able to interact directly with AI systems using only their thoughts. For example, systems that allow quadriplegics to type words and move a mouse cursor simply by thinking are already in use today.

These examples would represent a huge step forward from the current state of AI, and there are hurdles we’ll have to overcome to get there. We’ll need to optimize both the AI algorithms themselves and the infrastructure on which they run. It’s clear that many organizations have a long way to go in this regard. According to the GTTS, 42% of IT leaders said they’re not confident in their infrastructure’s ability to accommodate the growing use of AI.

Optimizing algorithms for trust and accuracy

Today’s AI architectures rely on analyzing massive volumes of data—an extremely resource-intensive process that would be difficult to maintain at scale. Instead of being data-driven, I expect future AI architectures will be data-informed. This means that they’ll be able to combine pattern recognition with human intuition. These models will be able to deliver more accurate results while consuming fewer compute cycles.

Another challenge is the fact that most humans are not able to explain how AI systems come up with a particular answer. This will need to change before AI adoption can increase. AI algorithms need to be explainable in order to be trusted. For instance, imagine what might happen if an AI system gives an inaccurate prediction that leads to a serious accident. If we can’t determine exactly what went wrong, it will lead to loss of trust in AI overall, rather than just that one faulty model.

I expect governments will play a role in establishing trusted AI. I think we’ll see tougher laws about what AI systems can and can’t do, and governments will design AI models of their own to help monitor and enforce those regulations. One area to watch will be data privacy and sovereignty regulations around AI. Today, there are few restrictions on how AI models consume data. I expect this will need to change before we can do AI at enterprise scale.

We’ll need data exchanges at both the individual and organizational level. These exchanges will ensure individuals have control over how, when and why AI models use their personal data. They’ll also track the lineage of models and data across organizational boundaries, so that AI-based solutions will always be able to demonstrate compliance with applicable regulations.

Optimizing infrastructure for efficiency and sustainability

Many industry observers have expressed concern about the carbon footprint of AI systems. For example, it took 1.287 gigawatt hours of power to train ChatGPT—roughly equivalent to the amount of power that 120 U.S. households consume over the course of a year—and that’s only one model out of many.[1] Going forward, we anticipate that many basic AI foundation models will become open sourced. We expect this will bring down the cost of AI pipelines (because people will not have to build AI models from scratch) and will also foster a lot of innovation.

A sustainable approach to AI requires algorithms that are more efficient, but we also anticipate we will need the digital infrastructure itself to be cleaner and more efficient. This will require using more renewable energy and taking advantage of sustainability innovations such as high-density liquid cooling.

I also anticipate that quantum computing will be fully mainstream in 25 years. This will enable wider adoption of AI, as quantum systems will solve certain types of AI problems much faster and more efficiently. Using a blend of quantum and conventional computers will help AI models consume less power—without having to sacrifice accuracy.

[1] Molly Loe, “Heating up: How much energy does AI use?” TechHQ, March 24, 2023.

I expect national governments will take a more hands-on approach to supporting the growth of digital infrastructure in the future, because they now recognize just how important it really is. Already, we’ve had several U.S. government agencies reach out to us to learn more about what we do, how we do it and why it’s important, and I can only assume others in the industry are experiencing the same thing.

One reason agencies want to learn more about digital infrastructure is that they know it will be essential to unlock the full potential of emerging technologies. Let’s take mobile networks as an example: In the years to come, we expect we’ll see the full rollout of 5G networks and the dawn of 6G. For these technologies to be fully realized, we’ll need a dramatic increase in data center capacity throughout the world and greater connectivity closer to digital edge locations. I expect policymakers will offer encouragement and guidance to help make this happen, because they’ll have a vested interest in growing those networks within their jurisdictions.

Legislators will push for greater sustainability and data privacy

I expect governments will support the growth of the digital infrastructure industry, but they’ll also want to see providers prioritize sustainability and data privacy as they grow. These issues will matter to governments because citizens will demand nothing less.

The data sovereignty movement is well underway in some regions, but it will become even more important as the world grows more interconnected. We believe nations need to fully participate in the global digital economy without placing their citizens’ privacy at risk, and that’s easier said than done. Governments will expect digital infrastructure providers to offer neutral platforms that make it quicker to deploy hybrid infrastructure in more places.

This hybrid infrastructure will make it easier for businesses to operate profitably in the global economy while also protecting sensitive data. This will fulfill a clear need in the business world today: In the GTTS, 56% of IT leaders said their organization is not well prepared to comply with stricter data sovereignty regulations due to insufficient infrastructure.

I expect Europe will lead the way for stricter sustainability regulations, just as it did for data privacy. The EU has already begun to release ambitious legislation that I think could become the de facto global standard. Digital infrastructure providers that already prioritize sustainability will thrive, while those that wait for government to force their hand could face serious challenges.

In 25 years, I believe it will be essentially impossible to find a data center that isn’t carbon neutral—especially in Europe. By then, the goal for the industry will have shifted to net-zero emissions, which will be a much more difficult undertaking. However, I think it will be a goal that both digital infrastructure providers and government regulators share. Rather than having an adversarial relationship, we’ll work in partnership to achieve what’s best for everyone involved.

I believe a business without a sustainability strategy is a business without a future. Many organizations still have work to do when it comes to creating a sustainability strategy that is focused on the right issues and driving the right actions. In the GTTS, less than half of IT leaders reported feeling fully confident that their organization can satisfy customer demand for more sustainable practices.

The bigger question is, what should that sustainability strategy look like? Broadly speaking, I believe the digital infrastructure industry needs to prioritize a global vision for sustainability, while designing and building locally and addressing regional material concerns. A sustainable approach to digital infrastructure will require a wide variety of material, technology and energy innovations, and not all of them are practical in all locations. Local climates, water availability and material availability are just a few considerations. Digital infrastructure providers must do what works, where it works, and then orchestrate all those different pieces into a holistic strategy.

A sustainability strategy is also inherently forward-looking, often with a timeframe well beyond typical business planning cycles. I think digital infrastructure providers will need to be ready to constantly adapt to changing infrastructure requirements and evolving strategies over the next 25 years. This includes potentially meeting increasing customer demands for efficiency, implementing new sustainability innovations as they become available and progressing to take on ever-bigger challenges as they arise.

Breaking through sustainability silos

Many businesses primarily focus on the sustainability of their own operations. They report on the progress they make with reducing emissions, increasing renewable energy use and other key metrics, but they don’t attempt to account for things happening outside their own four walls.

I do not believe this insular approach will be good enough to drive meaningful change over the next 25 years, and we are already seeing the need to address sustainability more broadly with increasing discussions around supply chain Scope 3 emissions reporting and requirements. Businesses are simply too interconnected with one another, and they’re only going to become more interconnected. In this blog, my colleagues have touched on how essential digital ecosystems will be in the future. I believe we’ll need an ecosystems-based approach to sustainability as well. This means not just evaluating your own sustainability metrics but also collaborating with customers, suppliers, partners and the local communities where you operate to drive sustainable outcomes.

One reason it’s so difficult for businesses to strategize outside their own silos is that it requires them to account for what their suppliers are doing, and their suppliers’ suppliers, and so on. We can’t directly control what our suppliers do, but we can steer them in the right direction. For instance, we can encourage them to set ambitious emissions-reduction targets by participating in the Science Based Targets initiative (“SBTi”).[2]

Participating in the circular economy

One way that enterprises can work together for a more sustainable future is to assess the full life cycle of their business processes.

For instance, when businesses deploy a new server today, they’ll likely evaluate the environmental impact based solely on the cost, performance and operations of the server itself. In the future, they should consider the entire server life cycle. They can do this by asking themselves:

• Does the server manufacturer have a sustainability strategy?
• What raw materials did the manufacturer use, and how were those materials extracted?
• What is the environmental impact of the data center where the server is hosted (including the embodied carbon of its building materials and its potential impact on local biodiversity)?
• Are there any environmental risks involved with the location where the server is hosted (such as drought, flooding or adverse weather)?
• What alternatives could they have considered instead of deploying a new server?
• Could they run the same server in a more efficient way or at higher capacity?
• What happens to the server at the end of its life cycle? Can refresh rates be extended? Can components be reused?

These questions are based on the concept of the circular economy. While a traditional linear economy operates on the principles of “take, make and waste,” a circular economy focuses on minimizing waste by reusing and recycling resources across every step of the business life cycle.

Everything we do as a digital infrastructure provider requires resources, whether it’s the materials that go into hardware or the energy that powers that hardware and cools the server halls. I believe it’s our responsibility to use these resources wisely, on behalf of our customers and the planet. In the future, I expect we’ll invest in new ways to optimize efficiency and minimize waste across the entire life cycle of our data centers and all the components that go into them.

[2] “About Us,” Science Based Targets initiative.

Prepare for a future defined by renewables

Another area where I think we’ll rely heavily on ecosystem partners is renewable energy production. Before we’ll be able to pursue net-zero emissions, I believe we’ll have to bring the supply of renewable energy closer to the demand. This won’t be easy, since not all renewable energy sources would be practical across all locations.

I predict energy companies will make significant technological breakthroughs over the next 25 years that will make it easier to scale up renewable energy production worldwide. No single technology will provide a silver bullet, though. Hydrogen, biogas and perhaps fusion—combined with renewables and distributed microgrids—may all play a part in the energy systems of the future. I think our role as a consumer of energy is to support the energy companies in their clean and renewable energy innovation efforts. One way we could do this is by demonstrating consistent demand for green electrons across our global footprint. Thus, our energy partners will have commercial support for them to invest in a greener energy infrastructure.

I also believe we need to maintain flexibility in our thinking across our operations. For instance, there’s limited availability of green hydrogen today, but I don’t think that will be the case in 25 years. We think we should begin to consider building adaptable sustainable design principles into our infrastructure today, so that as alternative sources of energy become available, we’ll be able to adapt quickly.

Equinix is putting our vision into action—today

As the digital world continues to grow more complex, interconnected and globally distributed, I am humbled by the fact that so many businesses continue to trust us to help them successfully navigate this changing world. I also know that as the world’s digital infrastructure company®, we’ll never stop working to be worthy of that trust.

To learn how we can help you prepare your business for the many changes ahead, read the Platform Equinix vision paper today.

Forward-Looking Statements

This blog contains forward-looking statements that involve risks and uncertainties as well as the personal opinions of certain members of our management. Actual results may differ materially from expectations discussed in such forward-looking statements. Factors that might cause such differences include, but are not limited to, risks to our business and operating results related to the current inflationary environment; foreign currency exchange rate fluctuations; increased costs to procure power and the general volatility in the global energy market; the challenges of acquiring, operating and constructing IBX® and xScale® data centers and developing, deploying and delivering Equinix products and solutions; unanticipated costs or difficulties relating to the integration of companies we have acquired or will acquire into Equinix; a failure to receive significant revenues from customers in recently built out or acquired data centers; failure to complete any financing arrangements contemplated from time to time; competition from existing and new competitors; the ability to generate sufficient cash flow or otherwise obtain funds to repay new or outstanding indebtedness; the loss or decline in business from our key customers; risks related to our taxation as a REIT and other risks described from time to time in Equinix filings with the Securities and Exchange Commission. In particular, see recent and upcoming Equinix quarterly and annual reports filed with the Securities and Exchange Commission, copies of which are available upon request from Equinix. Equinix does not assume any obligation to update the forward-looking information contained in this blog.

¹Molly Loe, “Heating up: How much energy does AI use?” TechHQ, March 24, 2023.
²“About Us,” Science Based Targets initiative.