IOWN: Putting vision into reality

A state of evolution

Assessing the proof points of the technology, Levrau makes a comparison with Darwin’s theory of evolution.

“Darwin was also mocked at the beginning, but [even] fibre optics were an experiment,” he says. “People saw the benefit when we [metaphorically] ‘came from the water’ into the lab, and it’s getting deeper and deeper. Why? Because [photonics] can be produced mass scale, and is energy efficient and has a higher capacity.

“I’m not going to say it’s easier to install, because photons are particular, but it has a lot of attributes that we have mastered over the years. Talking about the future, it’s clear that the capacity and scale our society demands – people talk about the ‘infinite connected society’ – can only be provided with optics or photons in general.”

But as with any technology, there are likely to be points of failure. Assessing how these are being avoided, Levrau draws on another analogy, likening the need to address legacy networking to a “huge metal ball in prison that prevents you from running very fast, but you need to take care of it”.

“The problem is that in introducing a new technology ... you always need to take the old [tech] into account,” he says. “Compare that to our society that wants to revolutionise the world, but forgets that we have an older generation to take care of too. There are failures [in older tech] that we need to take into account. We’ve accelerated our innovation on a tremendous scale to where we are now. My concern is whether it’s more of a society problem if we accelerate things too fast. Sometimes we want to run before we can properly crawl.”

In terms of the contribution of his company, Levrau says Nokia is contributing to leading with energy efficiency and in integrated mobile networks – it is working on proof of concepts – and on other tracks such as digital twins and fibre security.

“What’s important about developing the proof of concepts now is that success under the umbrella of IOWN is defined by proof of concepts, meaning you could work and write it all down, but if you can’t put the rubber to the road, what’s the value?” he adds.

Like Gonzalo Camarillo, head of implementation components at Ericsson and IOWN Global Forum marketing steering committee chair, Levrau cites artificial intelligence (AI) and digital twins as use cases that will have a particular impact on development.

“Digital twins for transport networks, or networks with a digital twin, is one thing that I will also participate in as a use case. The other one is training AI,” says Levrau. “We know AI needs a lot of traffic from the cloudification and virtualisation of infrastructure, for example, and the on-premise datacentres. Enterprises are now putting them in the cloud, which will put more strain on the network.

“Is it a new use case? No. It’s more of an evolution where you move data into the cloud. The newest one is the quantum computing state is typically maintained by spin of electrons, and for that you need a very fast connection.”

The technology maturity question

All of this assumes that the building blocks will be in place to support development. It’s worth looking at why now is the right time regarding the cost and value of current fundamental optical technology. Are these now mature enough for deployment?

Robert Keys, senior director of optical transmission at Ciena, believes the answer is very much yes, which is perhaps important given that the user focus of IOWN is aligned with what many of the technology firm’s customers are trying to do.

“Customers want the network to be open and they don’t want to be beholden to a single vendor,” says Keys. “They want to drive up the use of optics because that drives down space, power, and so on. They want to use those technologies together and to have as many suppliers in that space as possible. You can find a standard that people are generally happy to contribute to and that then allows you to build the network that you want. [Participating in IOWN] keeps us in touch with what our customers have asked from us.

“Think of submarine cables – they follow exactly that same set of principles. In the case of the submarine, you’ve got to have advanced technology to get that sort of capacity across 15,000km [for example]. But that same technology can be used to put them in routers that can connect distances that are not as challenging. Building the technology capability to address challenging environments naturally leads us to have the technologies that enable the more common, ‘mundane’ use cases.”

Solving the problem of going 50,000km means you’ve also solved the problem of going 100km, but the latter use cases will still have their own set of challenges. Broadly speaking, driving technology capability that improves the reach is regarded as vital for the types of users who will be attracted to what IOWN offers.

For Keys, the project had to make sure that Ciena could bring technology to the market within the context of a workforce and the way it wants to work – and this strikes at the heart of the technology maturity question.

“I come at this from a fundamental optical fibre perspective. What do I have there that was mature in that technology and what does that allow us to do? We have optical technologies on high-speed modems that can take us 15,000km, but they can give us 800Gb over 400, 500, 600km,” says Keys.

“We’ve also worked as an industry to define the operating modes of those modems such that they’re interoperable. You can buy a modem from Ciena and interoperate the modem from Cisco and Nokia. The industry has gotten together and said, ‘You can do this’.

“Models have come to the point where we can put 800 gigabit modems in a row, which coincides with capability of ours. So, that technology is mature, and by mature I mean any number of vendors can deliver it, so it’s not a unique technology. The optics that go on the line have been mature for decades, but it’s now much more programmable. The other thing we’ve done is to instrument the line so that it’s not blind anymore.

“We can monitor, measure and take that information  – again, via standard interfaces – and expose it to our customers, or they can read it directly in a standard way. Then you can make digital twins and predict things. You can tell customers the capability of their line and they can act on that. So, there’s a nice link between the fundamental technology, programmatic interfaces, the network management that consumes that, the process, and then feeding that back to the optimiser. It’s all there. It’s a single workflow right across the network.”

To an extent, elements of the network can be all photonic, from the access network through to a metro network, through the core and processes, and subsequently back. However, Keys notes that there are places where you probably don’t want it to be all photonic.

Looking at the application of IOWN technologies, Keys says the difficult part of developing tech is getting it to work in customer environments. This will be assisted, he predicts, by the fact that the IOWN ecosystem has insights on how the customer wants to set up a workflow. He adds that the project is like a village that’s developing a product that accelerates deployment and learning.

Measuring success

Few organisations are more attuned to user benefits than Sony Corporation. Speaking as a contributor to the IOWN Global Forum, Katsutoshi Itoh, head of connectivity technologies for the R&D Centre at Sony, noted that with compute-integrated network infrastructure, IOWN has the potential to be a platform for services relevant to the Sony business. Moreover, he believes it could form the fulcrum of the concept of a social infrastructure, addressing performance, sustainability and cost-effectiveness.

“Network performance now is a given condition. By engaging in early-stage, use case-driven specification work, we gain an opportunity to remove the constraint of network and compute performance specific to application to fully understand the capability of future infrastructure,” says Itoh.

“Early adoption use cases are the beginning of the evolution journey and never a single point of success measure. [These include] datacentre interconnect use cases which include enterprise site-to-site connection and are a very likely candidate…for the financial industry.

“In remote media live production for the broadcast industry, media production equipment can be seen as an application-specific datacentre, while remote GPU for AI [GPU to Enterprise] is an early adoption use case. NTT, NEC and Fujitsu are expected to play leading roles as technology suppliers.”

Looking at the differences between IOWN to other ventures, Itoh notes that the deployment of IOWN infrastructure has already started as an upgrade to the current infrastructure in Japan. While the market size of Japan itself will not justify the entire investment, flagship-type services and applications can start to address social needs. 

“Japan, being a ‘leader of ageing society, and de-population’, needs to also lead on social deployment of remote/automation societies,” says Itoh. “With many countries following ageing and de-population, leading an effort centred around IOWN technologies has the potential of spreading worldwide.”

Given all of this work and cooperation between NTT and its ecosystem, and firms such as Ericsson, Nokia and Ciena, what will the success objectives and criteria for IOWN look like?

Ciena’s Keys believes there is a fairly significant customer and ecosystem around customers to define what they want, and that companies such as his have brought the technology to the table so that they can achieve those objectives.

Ericsson’s Camarillo believes IOWN needs to define success in terms of deployment and running prototypes, stating: “[Technology just has to be] deployed. If you have deployed your technology and it is out there being used, it’s a success. Otherwise, it will [just be] nice exercises. It becomes ‘fine dining and good wines’. That would be my measure for success.”