Earlier this month, Australia-based cloud exchange provider Megaport partnered with Digital Realty to launch the Service Exchange cloud interconnection platform, with Megaport providing its SDN-based Ethernet fabric “on-ramps” to cloud providers worldwide – even the ones that don’t reside inside Digital Realty’s data centers. Megaport CEO Denver Maddux tells Disruptive.Asia editor John C Tanner why it’s a big deal, the importance of breaking cloud exchange services out of the walled-garden paradigm, and the challenges of doing that in APAC markets where older telco models still rule and data sovereignty matters.
Disruptive.Asia: Let’s start with the Digital Realty partnership and Service Exchange. What exactly is the goal there?
Denver Maddux: Service Exchange is their competitive answer to Equinix Internet Exchange and Equinix Cloud Exchange. It’s their envisioning of what a data center can do in terms of providing additional services to their customers.
Customers want to get access to more things – they don’t want to have to just go to this provider for this and that provider for that, and so on. If they can go to a trusted company like Digital Realty and say, “Hey I need some space and power and cross connect and I also want to get access to the cloud,” then they want to be able to offer that.
We’re doing the North American sites first. Then we’re rolling it out to a number of APAC and European sites after that, and then we’ll start planning out the next phase of sites toward the end of this month. They eventually want to tap this into every Digital Realty location in the world.
What specifically is Megaport bringing to that equation?
We have the largest cloud exchange on the planet today in terms of number of locations around the world, and we provide their Service Exchange that allows them to connect their customers to 132 data centers outside of what they built for Digital Realty. They want to be able to offer their customers a way to get access to other things and not just be stuck inside the four walls of Digital Realty, which ultimately makes it a better value play and a good reason for customers to put their money into it. So it’s not a walled garden, which is pretty disruptive. The fact that they’ve got their whole board behind that and whole executive leadership team behind it is pretty progressive. You don’t find that a lot in that market segment.
The unique aspect of our system is that we have a northbound API that any customer or partner has access to, which allows them to use software to send commands to Megaport and reconfigure the network elements that they’ve leased from us. You can change virtual services across that port or any other port that you have immediately rather than having to go through a lot of paperwork and waiting for an engineer somewhere to actually provision something.
With the cloud connectivity business getting increasingly crowded, what do you have to do to stand out from the competition?
Honestly, from our perspective it’s only competitive in the sense that someone else says, “I can get you to the cloud.” Beyond that, the entire conversation is totally different. Nobody actually gets people connected to the cloud at the speed and economics like what Megaport does. The closest thing to a competitor is frankly Equinix Cloud Exchange – which is only available on Equinix data centers. We run over 140 locations around the world that are spanning dozens of data center operators, so you’re not stuck with going to just one company for a contract to get access to those things.
Then you have our simple service delivery elements. Usually the longest pole in the tent for a customer to get access to Megaport is their cross connects from the data center. That takes anywhere from two to seven days. Once they’re on, then it’s software delivery for all of the other connections they want to make. If they want to connect to the cloud once they’ve turned up their Megaport, they can do that in under a minute. They can set up a connection from their infrastructure into any specific VPC in any of our supported cloud providers in under a minute. Nobody else can do that. And at the economics that we do it, there are usually several lesser zeros at the end of those numbers than our competitors as well.
Do customers get it when you first pitch this to them, or do they need some convincing?
It depends. The ones that get it are the ones that are already well down the path of a cloud option. Not all of them have adopted the concept of cloud networking, but many of them are moving to that, and if you can show them cost savings, or you can show them the flexibility involved, then they get that really quickly.
But not everyone is following that trend, so there are some people who are using the cloud but they’re just kind of doing some testbed stuff, fooling around and trying to figure out how it works, or they haven’t really strategized what it’s going to mean for their business. And then there are some that just haven’t gone to the cloud yet – they’ve deployed their own IT servers and they love what they’ve built, and they look at us and go, “Well, you’re just another Ethernet provider,” which we’re not.
Do you get that a lot in Asia-Pacific markets?
If you look at Hong Kong and Singapore, they’re still very telco-driven markets. So they’re very much focused on – and still absolutely okay with – a lot of their old telco models. In some ways, they look at our tech and go, “It’s cool, and if I find an application for it, I’ll use it.” And then we have to work hard at acquiring that customer.
What’s your biggest challenge in doing that?
One of the bigger challenges we’ve had in the market is name recognition. For a newcomer in these markets, it’s actually quite a slog. It can be very challenging to just get at the top of people’s minds, and also for them to trust you and realize that you’re investing into the market and you’re going to continue to grow here. So we spent a fair bit of time just getting the name out there, getting some good stuff and Hong Kong and Singapore and actually putting people local and markets for people to engage with and get to know. Those kinds of things are critical to markets in APAC because they’re still very relationship-driven, but it takes time for that sort of stuff to develop, whereas you go to some other markets and they can be very KPI-driven – “How much money are you saving me?” “How fast can you get me what I want?” – and then they move forward. But it’s a little different here.
For one thing, the concepts of data sovereignty are quite important to a lot of markets, and it’s played a lot into how people adopt and consume technologies like cloud. A few years ago, the investments into Hong Kong and Singapore, Australia and New Zealand, they were all very different from what they are today for cloud services. They had the same issues with South Korea and Japan – they’re very unique in their own ways, but all of them have that thing where they want to do something with the cloud but they don’t want their content sitting in another country, they want it sitting locally.
Which may be a legal requirement for them in any case.
Sure. And now, in the last five years there’s been a significant shift in terms of thinking about being local with cloud services. It takes a long time to actually pivot from something where the concept of cloud is that I can have my stuff anywhere on the planet – maybe you don’t exactly know where it’s sitting. But that just doesn’t cut it cut it for a lot of companies. It’s cool for test-dev folks that aren’t running any mission-critical apps. But once you start getting into the enterprise side of the business, which is where you want to get to and actually make money from, you better give them some confidence that you can show them exactly where their data is at, and that you can guarantee that it’s going to stay where it’s required to stay, and then you start to earn that trust.