Hello there, everyone. Welcome to episode number 679 of this here electronic engineering podcast called Amelia’s Weekly Fish Fry, brought to you by eejournal.com and written, produced, and hosted by yours truly, Amelia Dalton.

Folks, I just came back from the very best conference and expo I have ever attended — and that is saying a lot. I have been to a whole lot of conferences and expos over the years. I even spent most of my 29th birthday at the Flash Memory Conference!

And last week, I was invited to attend Upgrade 2026: Research to Reality, NTT’s annual conference, where NTT Research 2.0 takes center stage. I had the pleasure of chatting with several NTT associates, and today’s episode is the first installment of my coverage of this incredible show.

My guest today is Marco Provolo, Head of IOWN and NTT R&D Global Promotion Office.

Today, we’re diving deep into the future of connectivity, networking, and computing with a look at a revolutionary concept: the Innovative Optical and Wireless Network, or IOWN. This isn’t just an incremental upgrade — this is a fundamental shift, moving from electron-based communication to an all-optical infrastructure.

Marco and I discuss what IOWN is, why the shift from silicon to photonics is so crucial, how NTT is working with international partners to accelerate its adoption, and the most exciting applications we can expect to see in the near future.

So without further ado, please welcome Marco to Fish Fry.

---

Amelia: Hi Marco, thank you so much for joining me.

Marco: Thank you.

Amelia: Absolutely. Okay, so for my audience who may not know, talk to me about IOWN, and specifically the potential transformative impact that this next-generation communications infrastructure could have on the future of networking and computing.

Marco: First of all, IOWN stands for Innovative Optical and Wireless Network. It’s about creating a sort of revolution in communication and computing.

We understood that the amount of data — and the consumption related to data — has been growing over the past decades, and with AI, it’s only getting worse.

What NTT envisioned around 2018 and 2019 was the possibility of shifting from electronics to photonics. That means faster communication, more bandwidth, and less power consumption, which is probably one of the biggest challenges we have in computing today.

And when I say computing, I’m talking about everything connected with computers and the IT world, which is basically the foundation of our business.

The idea came after a couple of IPs and patents we developed in Tokyo with our R&D teams. We imagined the possibility of scaling this, but we knew we couldn’t do that alone.

That’s why we also created a consortium called the IOWN Global Forum. At the beginning of 2020, it counted basically three members: the founding members NTT, Intel, and Sony. Now it counts more than 170 members.

That means we’re trying to create a whole ecosystem around this because, again, we know we cannot do this alone. We cannot force some solution on the market — that’s never worked.

So the idea is basically to shift from electronics to photonics, and this brings a lot of advantages.

Going back to your question — why is this relevant for companies?

Sometimes, if you’re not in the business of data centers or IT, you might think you just have “the cloud” or a data center somewhere and things just work.

But cloud is not really a place. It’s someone else’s computer, and someone has to manage it. Geography matters a lot.

If you’re doing mission-critical operations and you’re latency-dependent — like in the financial sector — you don’t want to lose even a millisecond. You want to be as fast as possible.

That means the location of the data center has to be as close as possible to your data, and ultimately to where you perform operations.

If you’re a trading company, you need to be as close as possible to stock exchange servers in major cities. But basically, there’s no more space.

That creates a real problem for those who own infrastructure or need to leverage infrastructure.

Then we have another little problem — again, it’s called AI.

Five to ten years ago, the power consumption of a rack was around 5 kilowatts per rack. Now NVIDIA is selling systems that can require 500 kilowatts per rack.

In that case, you basically don’t have many options. You can either burn your data center to the ground, or free an entire floor for just one rack instead of hundreds.

Here, the possibility with IOWN is to connect computing — whatever that means. It might be one computer, it might be a micro data center, or it might be a huge data center — and connect that with IOWN to create infrastructure that, from the user perspective and from the data perspective, behaves as if it were one data center.

This creates huge advantages for expanding infrastructure and leveraging existing infrastructure, and that leads to many different business use cases.

What I’m currently talking about is APN. APN stands for All-Photonics Network. It’s only the networking part, but the roadmap of IOWN is much longer.

The whole process is about miniaturization.

Currently, we have transponders — that’s the word for the device that converts electricity into light and vice versa. The size is manageable; it’s called a pluggable. You can plug that into a switch.

That allows you to connect buildings and physical facilities.

The next thing coming to market by the end of this fiscal year — so by next April — is what we call PEC.

PEC stands for Photonic-Electronic Convergence devices. It’s basically the same thing I just explained, but smaller — small enough to plug in at the board level.

This means you’re going deeper into the data center, closer to computing.

Just to clarify, we’re not talking about photonic computing, which is another field. We’re talking about replacing as much copper as possible with optical fiber.

This is called PEC 2 because it’s the second evolution. We’re now very close to the chip, at the board level.

The third evolutionary step will be PEC 3, where we’ll be able to connect at the pin level of a chip — which means no more wires on the motherboard.

At that point, the question becomes: do I really need the motherboard?

Basically today, the problem of power consumption is very connected to bandwidth, distance, and data transfer rates.

That’s why all the components we have sit on the same board. Once you start moving things apart, power consumption increases dramatically.

With IOWN, that paradigm changes.

Imagine you have a switch here, and you’re not really concerned whether another chip, RAM, storage, or GPUs are right below it in a rack — or 50 kilometers away.

That is a very big change in how you think about infrastructure.

Currently, we talk a lot about edge cloud, edge computing, remote cloud, and remote locations.

What we expect in the next five years is that those distinctions become blurred, because you’re no longer really interested in where your computing is physically located.

Actually, one interesting concept is that today we think about infrastructure as computing infrastructure. We tend to move data to where computing is.

So if you have GPUs for training models or inferencing, you try to keep data as close as possible to those GPUs for performance reasons.

But if you think about having infrastructure that is basically dematerialized — where location matters less — then you can think about infrastructure based on where your data is, not where your computing is.

That leads to very interesting conversations around data sovereignty.

---

Amelia: All right. So Marco, let’s also talk about how you and the NTT R&D Global Promotion Office are helping accelerate the adoption of IOWN, specifically through collaboration with international partners and research institutions.

Marco: That goes back to what I explained before — the IOWN Global Forum.

That is a huge resource for us.

In there, we have telcos, network equipment vendors, research institutions, potential clients, hyperscalers, and hardware manufacturers.

NTT is actually playing a big role because sometimes we’re a telco, sometimes we’re a system integrator, and sometimes we’re a consulting company.

The big advantage we have is that we talk constantly with our clients.

Sometimes solutions appear in unexpected places simply because our clients know their business better than we do.

What we offer here is a tool — but clients know exactly how to use it.

A couple of examples are pretty interesting.

The first one is banking and financial institutions.

Broadly speaking, they rely heavily on private data centers and have a huge number of use cases to solve.

As I mentioned before, companies today think about infrastructure as computing-driven infrastructure.

In tiny and fragmented Europe, when you generate data in different countries, you often need to keep that data in those countries.

And in Central Europe, every 50 kilometers, you’re in a different country.

That means if you need to manage infrastructure, you have to distribute GPUs, memory, and storage everywhere.

The advantage with IOWN is that you don’t need to care so much about where your computing is located. You focus on the data.

That’s something you can do gradually and evolutionarily.

There are also situations where you simply couldn’t do certain use cases before.

For example, we have strong relationships with mining companies.

The future vision of mining is to avoid having people physically working in mines because, as you know, it’s dangerous, dirty, remote, and difficult.

You can’t stay there for long shifts.

So the future vision is a fully autonomous mine.

The problem is that you cannot shift to full autonomy immediately.

The middle step is remote operations.

We can already provide remote operations, but companies face massive problems with bandwidth and latency, which directly limits scalability.

We can operate one digging machine, one excavator — but not 100.

This is where IOWN technology can really shine.

The first reason is straightforward: guaranteed super-high speed, very low latency, and very high bandwidth.

But there’s another important point — resiliency.

Mining environments are difficult. Sometimes you get what they call “micro-cuts,” where the connection becomes unstable.

That’s because of the harsh environment.

One side advantage of IOWN is resiliency. You can create a ring of optical fiber connections and switch very rapidly between one data center and another, or one operation point and another.

That means you can always guarantee high-quality connection no matter which fiber path you’re using.

That’s crucial.

You cannot remotely operate heavy machinery and then suddenly lose connection for a few seconds. That creates serious operational danger.

It’s not really a technological problem — it’s a connectivity problem.

Another interesting application is in products.

Twenty years ago, products were just physical products.

Then IoT came in, and we started adding services to physical things. Bluetooth and Wi-Fi were more than enough.

Now we’re seeing a lot of possibilities with AI.

But you can’t put a GPU in every device. That’s too costly and creates power consumption problems.

So one way to scale intelligence is through the network.

Instead of putting a GPU inside the device, you connect it externally.

One example is an endoscope.

Doctors need augmented reality information generated by AI for diagnosis — AI-aided diagnosis.

Initially, companies embedded AI directly into the device, but that was very expensive.

What we did instead was connect the device to a public cloud — this has already been done in Japan.

So you can still use the older “stupid” device, but connect it to a GPU cluster located 150 kilometers away and still get immediate feedback.

The doctor can move the probe and get instant AI-assisted analysis.

This isn’t just about creating smarter products — it’s about making intelligence scalable.

---

Amelia: Sure, that makes sense. So what do you think are the primary technical and business challenges you’ve encountered in promoting and implementing IOWN technologies, and how are you working to overcome them?

Marco: This is not really about introducing entirely new technology.

It’s a lot of existing technologies combined in a new way.

Our R&D is public, so we don’t own the IP in a closed sense. That goes in the direction of the consortium.

The idea is to create a new ecosystem that delivers clear advantages.

The challenge is that we’re not bringing some magical new invention from Mars or the Moon.

We’re applying known technologies differently, in a new architectural pattern that creates valuable solutions.

That’s the challenge.

We’re trying to change the way people think about connectivity.

Today, connectivity is mainly managed by telcos.

Our goal is to democratize fiber.

We want companies to understand that they can manage their own networks — or we as NTT Data can manage it for them — and they can achieve stronger KPIs and better performance by doing so.

---

Amelia: What do you see as the most compelling use cases or applications emerging from this architecture in the next three to five years?

Marco: The big advantage is that data centers are becoming a real problem — for citizens, for companies, and for countries trying to figure out where to place them.

In large countries like the U.S., you potentially have space.

But in smaller, denser countries like those in Europe, Singapore, or Japan, placing data centers is much harder.

The advantage here is finding a solution to that.

Also, being able to manage your data and define where your data resides — without worrying about where the computation happens — is probably the biggest shift we’ll see in the coming years.

This is both an advantage and a challenge.

But I think we’re on the right path because of our constant conversations with clients.

Sometimes we offer connectivity, sometimes services, sometimes consulting — but the backbone is always the network.

The real goal of something like IOWN should be commoditization.

You shouldn’t need to think about how the network works.

You should only care about how your business runs.

---

Amelia: All right, Marco, it is time for your off-the-cuff question.

If you could have one meal right now — it doesn’t matter if it’s on the other side of the world and you need a passport to get there — what would you have?

Marco: I think I would have lasagna.

That’s very straightforward, being Italian.

I’d love the kind of Easter lunch, Christmas lunch, or Sunday lunch experience — lasagna and some…

I’m actually vegetarian now, but if I had to choose, I’d go for some really good stewed meat.

---

Amelia: I love it. Awesome.

Well, it was a pleasure speaking with you, Marco. Thank you so much for joining me.

Marco: Thank you very much for this interview. Much appreciated.

---

If you would like even more information about this topic or Upgrade 2026, I’ve included a couple of links below the player on this week’s Fish Fry page on eejournal.com and in the description for this week’s YouTube episode as well.

Hey — have you checked out EE Journal on social media yet? Well, you should!

You can find us at facebook.com/eejournal.

If LinkedIn is your thing — I dig it — you can follow me or us on LinkedIn.

And we are also on BlueSky Social and Mastodon, too.

And we have a YouTube channel: youtube.com/eejournal.

Folks, it is chock-full of all kinds of techie videos, including our very popular Chalk Talk webcast series hosted by me.

And of course, you can subscribe to our EE Journal YouTube channel as well.

Thank you everyone for tuning in.

If you know of any cool new technology — or heck, you just want to chat — shoot me a line at Amelia, that’s AMELIA@eejournal.com, or post a comment on our forums at EEJournal.

For the week of April 24, 2026, I’m Amelia Dalton, and you’ve been fried!