xiaoliangge - stock.adobe.com

IOWN: Shining light on the future of communications

With rising demand for data and energy consumption due to the vast compute power required by applications such as AI and large language models, something needs to change in networks – this could be IOWN

The year 2030 is set to be momentous for those in the world of communications. If development work keeps to its current pace, it will be the year the world sees the advent of fully fledged commercial 6G networks that promise to be transformational on both a technological and societal basis.

At the same time, and perhaps in the case of light being kept underground as opposed to under a bushel, there will also likely be a revolution in fixed communications, specifically through the Innovative Optical and Wireless Network (IOWN) project coming to fruition.

Spearheaded by comms and IT giant NTT – and supported by Sony, Intel, Nvidia, Microsoft and other leading technology firms – the IOWN initiative envisions a future global communications infrastructure capable of enabling ultra-high-speed, high-capacity internet services utilising photonics-based technologies.

It aims to address the almost exponentially rising demand for data, and a commensurate rise in energy consumption due to the vast amounts of compute power required by future applications, in particular artificial intelligence (AI) and large language model (LLM) use cases. This network and information processing infrastructure includes terminals that can provide high-speed, high-capacity communication using technology focused on optics, as well as large computational resources.

Speed-of-light mission

In its prospectus, IOWN says its core mission is “turning science fiction into science fact”, showing how speed-of-light technologies will improve daily life, work and society at large. From a technological basis, IOWN comprises three major technical fields: an all-photonics network (APN), digital twin computing, and a cognitive foundation.

The APN’s photonics and optics-based technologies are designed to achieve three performance targets: ultra-high capacity with data processing of 125 times greater than current networks by volume; ultra-low latency offering near-instant transmissions with end-to-end latency reduced by over 200 times; and ultra-low power consumption with a goal of 100 times more efficiency than current transmissions, reducing carbon emissions by 45%.

NTT has also committed to promoting the realisation of a new communications platform consisting of the APN, which includes silicon photonics, edge computing and wireless distributed computing, through the development of new technologies, frameworks, technical specifications and designs. Testing of NTT’s APN in datacentres began in Tokyo in 2023 in partnership with Amazon Web Services (AWS).

IOWN's core mission is “turning science fiction into science fact”, showing how speed-of-light technologies will improve daily life, work and society at large

The digital twin computing element is an extension of conventional digital twins that uses photonics-based computational capabilities to perform calculations on virtual models of objects and humans to accurately make predictions about the future. The Cognitive Foundation forms an optimised, autonomous control of tech resources, including cloud, edge, networks and terminals based on artificial intelligence and machine learning.

In a 2023 report looking at the prospective emerging candidate technologies that would underpin IOWN, NTT highlighted photonic-ready, distributed datacentres and IOWN for AI and LLMs. It also showed the initiative’s goal of 100 times more efficient transmissions. Additionally, it mapped out the future deployment of APN technologies, whose low-latency capabilities will enable a shift from centralised datacentres to distributed datacentre “stations”.

The datacentre work is, said NTT, necessitated by rising data demand and the related rise in energy consumption, expected to exceed 3,000TWh worldwide by 2030. In the domain of AI and LLMs, NTT noted that these areas require massive amounts of compute power, calculating the energy required to train an LLM once is equivalent to that needed to power a nuclear plant for an hour. IOWN’s ultra-high-capacity goals look to address these issues.

Additionally, NTT announced its own proprietary LLM, Tsuzumi, an ultra-lightweight model to reduce learning and inference costs.

The report also highlighted growing AI use cases for IOWN-based networking. These included AI-based diagnostics, low-latency remote surgeries and the further development of bio-digital twins.

Concrete roadmap and industry buy-in

To realise IOWN, NTT said it was necessary to craft a “concrete” technology roadmap and promote technological development. Once drawn, the company said that would accelerate the incorporation of data-centric and disaggregated computing technology into IOWN.

Also necessary is to gain buy-in for its plans from the wider technology ecosystem. To this end, NTT and other key partners have established the IOWN Global Forum (IOWN GF). The forum recently convened in Vancouver in its first-ever public meeting to reveal not just the progress made in the project, and the likely technological advances, but also how the ecosystem is growing and what IOWN means to the key stakeholders.

Among the insights offered into the mission and work of IOWN GF, the sessions – attended by around 200 delegates from over 140 companies – showed the fundamental vision of the comms world in 2030; how to prepare for a more secure and sustainable world, powered by light; how open source collaboration would be key to driving the future of communications infrastructures; and the possibilities of how IOWN technologies will likely emerge and what they could achieve in this timeframe.

In explaining its mission, IOWN GF is keen to emphasise that it’s not just “any” technology groups and technology projects forum, which those in the tech industry may be used to. It stresses that among the reasons why it is different from other groups of its kind is that fundamentally it is real deployment and use case oriented and goes direct to users with the clear aim of bringing optical networking – and the domain of computing – to the masses.

Mission, vision, focus

Speaking to Computer Weekly, Gonzalo Camarillo, head of implementation components at Ericsson and IOWN GF marketing steering committee chair, says a key message that the forum needs to drive to the industry is that IOWN is “not selling smoke” to a communications ecosystem that is well used to alliances and trade associations with regular advances in technology. Instead, he insists that IOWN GF will show how to create value from the technologies that are being developed and take advantage of the sheer diversity of the membership when looking at the three pillars of mission, vision and focus.

“We are, of course, very strong in Japan and APAC, but we also have strong members from the US and Europe. We have Ericsson Nokia, Orange, Telefónica, Microsoft, Intel, Google, etc. So in a sense, geographically we are covering [large areas], and that’s why in our messaging framework, we stress that if we want to make [IOWN GF] happen it cannot be Japanese-only or APAC-only. It needs to be global,” says Camarillo.

We have a vision for 2030, but the timeline starts next year. In 2025, we’re going to have the first early deployments
Gonzalo Camarillo, Ericsson and IOWN GF

“On the other axis, where we are also end to end, is industry verticals. We have people from the banking industry, the finance industry, the broadcasting industry, TV entertainment, the telecoms industry and automotive. And then the third axis, where we also have diversity, is the role a member plays in its industry. That is the entities that deploy the technology, the operators of the technology, equipment vendors. We even have people doing applied research.”

Yet the focus going forward will be applying these pillars of operation with those that are going to be using technologies developed under the IOWN umbrella. Camarillo suggests that a bank could be using an IOWN network for their communications, or Volvo in manufacturing cars and supporting its associated connected multi-disciplinary manufacturing ecosystem. Such companies will, before too long, be looking at real use cases as they put together requirements specifications.

“They will say, ‘Look, I need to connect these two datacentres’, or, ‘I need to perform these financial transactions and these are the issues I’m facing, and this is kind of what I would like to see’. So with those use cases, we can analyse them or take out and extract the requirements. And then we can start working on the technology that would address those requirements,” he says. “Once we have the requirements, [with] blueprints and architectures and all that, we can move onto proof of concepts to prove that the technologies we are developing make sense. And after that, we move into early deployment. We have a vision for 2030, but the timeline starts next year. In 2025, we’re going to have the first early deployments.”

Especially because his company is renowned for its leadership in communications technology, Camarillo is also keen to point out that the development roadmap of IOWN and that of the forthcoming 6G ecosystem are really coincidental, even though both promise to be as transformational and create smarter societies.

He also expresses confidence that IOWN GF has an ever-clearer picture on what it is trying to achieve, in the face of the ever-increasing data explosion, something that has come to particular prominence in Japan where AI adoption is seeing data growing. This presents challenges that cannot be shirked, as AI is seen as vital for the future prospects of the country’s economy as a whole.

“Data flowing through the internet is growing exponentially in Tokyo mobile networks. They are handling double the amount as they did before Covid. There is a [key] macroeconomic element about Japan. AI is seen as a solution to skills crises, meaning AI is driving the consumption of data big time,” says Camarillo.

We are starting to work on focus use cases ,which with AI [involves] GPUs and how to how to handle data growth. We cannot be honestly [thinking] of the network of the future without thinking of AI. AI is everywhere right now. It is arguably the hottest technology. We are [also] trying to explain that our networks are going to be more secure. Optical computing will be much faster and allow types of encryption and protection that are not possible today.”

In part two of our look at the work of the IOWN project, we find out what use cases the new infrastructure will support.

Read more about optical networks

Read more on Telecoms networks and broadband communications

Data Center
Data Management