Leading the Charge on Data Gravity with Digital Realty’s Ben Gonyea

Today’s guest on Good Morning Work From Home is Ben Gonyea, VP Global Product Management and Development, Digital Realty. Ben joins us to discuss how Digital Realty is leading the pack with the digital gravity transformation, and how leveraging solutions like Platform Digital can enable data centers to be the hub and spoke to better manage rapid change.

COVID-19 hit everybody and has lasted longer than I think most of us expected. How has that impacted business for Digital Realty? How’s your business helping people? 

For us here at Digital Realty, we’ve fortunately been busier than we normally would have been. What COVID has done is basically escalated the digitization of businesses. A lot of companies had roadmaps and plans spanning three to five years that basically got pushed to being done in 30 to 60 days. A lot of our end customers have been adding capacity to try to keep up, as well as support new types of business, specifically online learning, meetings, conferences, and all of the support businesses around that. They’re adding capacity and generally leveraging the internet to do more.

So for us, we’re probably one of the handful of industries that have been busier and indirectly benefited from this situation. What we see now is just going to continue moving forward, and a lot of the migration from legacy in-house, data centers, or stuff that was done face to face will be virtual.

The one term that I keep hearing when I hear about Digital Realty is the term data gravity. So what does data gravity mean to your team and how is Digital Realty in the center of the gravitational pull there?

Data gravity is really the movement of that data that I highlighted before. We’re really focused on the enterprise data. That is that data that’s still potentially sitting in corporate-owned data centers or was done on an intranet within an office environment before that’s now being pushed into the internet or between regions, within North America or across continents potentially. What we’re really focused on is being able to help businesses leverage their data gravity right — basically create another ecosystem. Years ago, it was done around the carrier center kind of model. Now it’s really about the data being generated by enterprises. That can be from the enterprise side of things that’s being driven from even households or businesses and needing to move that data around. It can be the transactional nature of business that’s moved online and it’s really leveraging our data centers to be the hub and spoke of that model.

That’s what we’re doing from a product standpoint with Platform Digital so we’ve created use cases around that. We have three hubs, in essence, that we built looking at our very large customer base. What we’ve done is we’ve created these use cases to talk to other enterprises about what we feel is really best in practice and then we work with them to try to find the best-suited applications for them. What we’ve done is build a network hub, which is something traditionally we’ve had in our data centers. The evolution is that the enterprises are now managing more of their network directly. And they want to be hands-on with that because when you start talking multi-cloud or private and public cloud –– they want to control that network interface.

Then, you basically layer on what was traditionally a lot of the traditional hub and spoke model of their networks, connecting their different enterprise locations and now it just becomes a broader network. So the network hub kind of sits in the middle of that. Then, we have a control hub, which is really that next layer up. When you start layering on managed services from our partners’ security applications, as an example, you then can put that in a data center. And not only do you leverage those applications for whatever is data center-centric in the environment, but then across potentially your entire network, that’s the network hub is sitting there with adjacency. Then the last one is the data hub. The data hub is what we’ve seen moving quicker than it was before, moving those traditional data warehouses out of those corporate data centers and into an environment like ours.

What that really does is allow them to potentially move a portion of their data into cloud environments or expand up to accommodate for things like gaming at certain times of day or periods of gaming. If you’re an online retailer, you can expand up at certain times of the year, which gives them more flexibility managing their holistic data needs. We’ve really seen great adoption with that. The other thing that we’ve seen is that it brings on a very collaborative working dynamic between us and the customer. We can say to the customer “Here’s your use case — this is what we’ve seen, we’ve analyzed hundreds of our customers, and this is what it looks like. Now let’s hear about what you’re doing in your applications.”. And then we find the solution best suited for them.

I’d say what we’re seeing out there is that data is the new oil, right? When you start looking at enterprises and companies, they really are trying to figure out ways to mine and get more insights out of data and to do so they’re storing a ton of data — a lot more than they would have in the past — or gathering more data from more devices and more endpoints. Platform Digital and that whole concept of data gravity puts Digital Realty right in the center of really enabling customers to leverage that data in ways that they couldn’t before.

Absolutely. To your point, it IS kind of like the new oil, because what our customers have realized and shared with us is that by looking at their data and cutting it in different ways, they’re able to make better decisions, quantitatively, where before they might’ve been qualitative decisions. They go through it with their customers and they get insight. They might go down and offer a new product offering, or they might modify the way that they’re leveraging use even at our own data centers. When you look at a decent application or our data center monitoring, or you’re looking at our infrastructure, you know, what are the setpoints? How is airflow going, right? That comes down to if it can save you off X. If you’re looking at that data more frequently, if you look at it on a quarterly basis, or you look at it monthly, you can make certain assumptions.

You can make certain changes. If you get closer and closer to real-time, the monetary impact is greater, right? Every tweak that you make, you’re in essence improving your OPEX. So that’s like a real life example kind of from our own environments, but any of our customers, regardless if it’s a painting company or an oil company, you just want to optimize your data to figure out where to best put your investment capital. Being able to leverage cloud resources for that means that you can then analyze your data when you need to versus having to basically do it in house. You might not need to analyze your data all the time, but when you do need to do it, you can look from a variety of partners to get that information and leverage that type of application service.

If we could tie that back into Platform Digital, if I’m understanding you correctly, what Digital Realty has created is really a set of reference architecture that you’ve collected from some of your largest customers that have best practices they’ve been doing for a long time. You’ve got that reference architecture so that people that are kind of new to the data center environment, to the networking environment that controls what equipment is best, how to set that equipment up, how much power is going to draw, how much cooling is going to need, where to put it understand what has to happen. It’s like you’ve created a bunch of templates so that people who may be new to it can avoid making the mistakes that they may have made on their own. And you’ve really helped them accelerate their business. Is that the right way to look at it?

You’ve captured it perfectly! What we did is we analyzed our global 2000 customers that we have from the enterprise side. We looked at how they were doing it. What are the applications they’re running? What’s the infrastructure they’re putting in? What types of network interfaces are they doing?

Then, we go with reference architecture to other enterprise clients and then say, this is what we see as best in class from a reference market. This leverages the light types of power density. This reference is the right type of security. Then we start the dialogue because every customer is going to have a unique set of needs, but it gives them an area that they might have not thought of, or it might be something that’s on their roadmap if they pull it in.

When you look at a total cost of ownership type of model, they might be able to do a lot more than they thought on day one, like they’re providing a portion of their infrastructure to a data center. They might be able to move, not only the physical infrastructure but their cloud instances and then their network interfaces. We start with those reference architectures and start working in and basically bridge a dialogue to then create a best in class type of solution.

Anytime we’re coming up on an election, there’s a big discussion around the environment, the carbon footprint and green energy. More power than ever is being consumed by gara centers. What’s Digital Realty’s take on green energy? Where is the company headed, and how are you helping customers meet their own goals in this area?  

First, I’ll start outside the data centers. In my experience, a lot of the corporate enterprise data centers are underutilized. They built a facility forecasted for eight megawatts of use, but they’re currently using two. When you look at a data center like that, you still have to run the infrastructure — you have to run a certain portion of it and you can’t shut it all off. So your efficiency of how you’re cooling just drives up your overall energy use of the entire building. When you look at optimizing with someone like ourselves, we’re putting in multiple customers and if you look at a co-location environment into the same footprint, our whole goal is to optimize the power. When we get power allocation from a power company, based on how it’s allocated, we need to optimize that or we can’t get the next allocation. We have to do it efficiently.

In terms of improving the efficiency of power, when they move into an environment like ours, once they get to our data center, there’s a couple of different areas that we’re very focused on from a green perspective. The first is just our design. A few years ago we moved to a distributed redundant design from a twin topology, same resiliency as before, but it did save on the PUE. Overall, it also saved on the amount of infrastructure that we needed across the board, which then also reduced our energy need and the footprint of the infrastructure and what it took up in a data center.

Then when we get inside of the space, we’re into basically design standard requirements are our decks and what customers have to do when they deploy. So on the scale side, we help them optimize their footprint, right? We provide power density. We look at how they can put the configuration in to use the power as well as the cooling most efficiently. Then, we look at the co-location side because we control a lot more of that. We put requirements in place and for how we’re going to do containment or how we’re going to put in blanking panels, creating that lower POE from the get-go. Then, when we look at it from a holistic standpoint. We purchased a wind energy farm in Texas and we put it on the grid. When we do enough of that it offsets our North American co-location needs. That’s just one way we’re trying to help be greener.

In talking about our scale and hyperscale clients, if they’re looking for green energy, we have a group that helps them source that based on the region they’re in and what types of green energy is available. Is it wind? Is it solar? Is it a combination? We have groups that will help them try to find their green goals when sourcing green power and being a large purchaser globally. We can leverage a lot of our abilities to source power because we show demand. It’s a lot harder to get committed power when you don’t have historical demand, but because we’ve had data centers for decades we’re able to basically show the consistency — and that’s one good thing about data centers. Our power usage is a pretty steady state and it changes seasonally. Typically you don’t see ups and downs and spikes. So that means we’re able to source the power in a consistent manner and not have to go on and purchase it based on fluctuations.

Do you need help with your next technology project? LET’S TALK



Experience the Bridgepointe Way

Start today with a no-obligation consultation with one of our experts.

Table of Contents

Related Blog Posts