Network architectures and carrier strategies are evolving towards a platform play model. Research company IDC’s Hugh Ujhazy tells Bill Bennett the future will see connectivity aligning with business outcomes and transport mechanics becoming less relevant. Connectivity strategies will be driven more by the desired outcome and less by the journey

Wide Area Networking is evolving at a pace. WAN users already combine fixed-line and wireless technologies with concepts like software-defined networking.

This approach is better suited to cope with a world where workers, applications and workloads are in a constant state of flux.

Eventually we could reach a point where just as cloud computing users don’t always need to know where their data is, network users won’t need to know how their data gets from point A to B. All they will care about is that it gets there on time and at a reasonable cost. 

IDC vice-president Hugh Ujhazy says a year ago he was talking about carrier supported multi-LAN (Local Area Network) and multi-cloud capabilities. “There is an explosion of options along with the race to interconnect all the public cloud, private cloud and on-premises clouds, and wrap all this into the overall enterprise network.

“This year we’re adding the concept of multi-platforms because we think the future is moving towards a platform play. This means an application will make a request for a connection, it will give the request some parameters and it will then do whatever job it was going to do before handing the connection back. This is in contrast to the conventional approach where a customer might have to get a fixed two-year contract for a link.”

With a platform play, telcos will offer connectivity services, alongside compute, storage and applications, on terms that are similar to cloud computing. Just as cloud customers don’t have to pay for on-premises servers but spin up on-demand services instead, connectivity customers won’t need fixed links. They’ll be able to buy the connections they need as they need them. And, like the computing cloud, the network will be always on.

Ujhazy says: “We have to assume the network is always going to be. You might make a request such as, ‘I want connectivity to an end point in Hong Kong. I want one millisecond latency and I want the least cost route’.”

This is a way of abstracting communications to the point where customers don’t need to know or care about what is going on. In effect, they are asking for a connection, letting the network operator choose the technology and the route. All they do is pay for the delivered service. There is no need for a permanent relationship with the network.

Ujhazy likens this to loading an app on a phone. He says: “It goes out there, grabs some cloudy bits, some application bits and does its thing.”

We are not there yet. There is still a way to go, says Ujhazy. The message from the carriers is: “If we present ourselves as a utility where all I’m offering you is a pipe to shove data down then I’m increasingly marginalised. So, if I start offering you a connectivity platform which has a place to store all this data you are connecting, a place to consume it from any device from any location at any time, and a place to share it, then I can become really relevant.”

He calls this approach “moving more and more behind the curtain. Think about what AWS (Amazon Web Services) and Microsoft Azure have done for us. There was a time when we had to work for a living while building these environments. Now it’s a matter of matching this much virtual machine with this much storage and this much capacity and here is the credit card.


Ujhazy says he is impressed at what Singtel is doing in terms of unifying all these parts. He says: “They’ve built an interface on top of all the cloud providers and all this connectivity. It’s called the Liquid Infrastructure. Customers can come in through the Liquid Portal and they can choose this and that and it comes down to a single Singtel bill. That’s the sort of complexity that we can push behind the curtain and make it the responsibility of the carrier. It’s quite appealing to customers who don’t want to deal with all that.”

Communications technologies are converging elsewhere. Ujhazy says that in some markets Vodafone plans to offer “IoT Plus”. This brings together traditional IoT (Internet of Things) connectivity with private LTE (Long Term Evolution) and 5G networks, along with edge computing. This is all wrapped together with managed services. In effect, Vodafone is telling customers it will take care of everything. It’s a compelling proposition.


Another aspect of 5G is that it won’t be an overnight revolution. He says: “It is going to be with us for the next 20 years. There’s no need to run up to the buffet to grab a quick bite, it will be here for a while. 

“Everything about the landscape for 5G is a co-existence strategy with 4G. And that 4G network isn’t going anywhere probably for the next two decades. When we look at things like broad widescale coverage over long distances, a lot of that is best left to 4G.” 

The big change will come with stand-alone 5G. Once the millimetre spectrum is available, that will be where customers get to see the promised high speeds. He says: “It is only going to be there in pockets where it is needed. You’ll see a blended environment in terms of the Gs for some time.”

This mixed environment goes for all the Gs: 3, 4 and 5G. Ujhazy says each of them does something interesting. “The 3G is used today for M2M (machine to machine), for cellular connected devices that will probably move to narrowband. Some markets are retiring their 3G, but the message here is all about the overall portfolio of connectivity options you now have as a user.

“It’s the same for enterprise customers and consumers. You have everything from fibre to Wi-Fi 6 to 4G and 5G. We’ve got more choices than ever before and building the right strategy for that is the challenge we face.”


Ujhazy says 5G is going to be huge for the mobile carriers.

He says: “We’ve done a telco capex (capital expenditure) forecast for the Asia-Pacific region excluding China and Japan. The fastest growing single capex item is the build out of 5G radio access networks. It’s going to grow at 93 percent compound annual growth rate (CAGR) over the next five years. That’s in a context of a decline of about 1.6 percent CAGR in their total capex. So, it’s obviously going to be big for operators.”

For Ujhazy, 5G remains mainly an enterprise play, but there is also a consumer play in the mix. He says carriers need 5G technology because it simplifies and streamlines their job of provisioning and managing a cellular customer. They also need it so they can keep offering the increase in performance and capacity that we’ve all come to expect.

He has changed his view of network slicing in the last year. “I didn’t believe in the private networks that are possible with 5G. Vodafone Global changed my mind. They said they have about 160 different requirements – that’s expressions of customer interest – already on the table. So there will be a place for it. I’m not sure how that will make sense over time. But it is coming.”


Sydney-based Hugh Ujhazy is vice president, IoT and telecommunications for IDC. He leads the research company’s analysis of fixed and mobile network services for the Asia-Pacific region.