The plan to make all networks optical is about to take two big steps forward
(2025/08/07)
- Reference: 1754563510
- News link: https://www.theregister.co.uk/2025/08/07/iown/
- Source link:
In December 2024, Japanese tech giant NTT revealed two impressive feats of high-speed networking.
Working with Hitachi Vantara, the company [1]virtualized a pair of storage arrays into a single logical unit that achieved real-time data synchronization. That's a neat trick in the datacenter, but NTT and Hitachi pulled it off despite the two arrays being in different facilities separated by 600 km – the kind of distance that creates lots of latency and makes real-time transfers impossible.
Later in the month, NTT [2]detailed its efforts to shift data at an astounding 455 terabits per second across 1,000 km using ordinary optic fibers.
[3]
The common denominator in both tests was technology called "Innovative Optical and Wireless Network" (IOWN), an all-optical networking stack that NTT hopes will mature in 2030 and expects will reduce power consumption by 100x, improve transmission capacity by 125x, and reduce network latency to 0.5 percent of current levels.
[4]
[5]
IOWN draws on research into optical transistors conducted by NTT boffins, whose work made it into the Nature Photonics journal [6]in 2019 .
Transistors are electronic devices, and signals that pass through electronic systems degrade due to electromagnetic interference. Optical communications, by contrast, are largely immune to electromagnetic interference, and can carry data faster and more reliably over long distances.
[7]
But signals carried on optical networks eventually reach electronic devices – and when that happens, the data flow slows.
Through IOWN, NTT plans to stretch optical networks deeper into computing infrastructure. Instead of data sprinting across an optical WAN before strolling through an electronic router and crawling across copper, NTT wants information to spend more time on optical media and switch through optical transistors, and flow at high speed for longer.
NTT created the [8]IOWN Forum to explore and advance such scenarios, and some of the biggest names in IT have joined it: Google, Microsoft, Cisco, Dell, Qualcomm, Intel, Nvidia, and Broadcom are all aboard. Japanese industrial giants like Mitsubishi have signed up. Telco giants BT and Telefonica are members, as are Ericsson and Nokia. Sony, which knows a thing or two about optics after working on Blu-Ray, is contributing research. The International Telecommunication Union last week [9]agreed "to accelerate further collaboration."
From WAN to motherboard
The IOWN Forum has set what it calls "lighthouse targets" to lower network power consumption by 100x, improve transmission capacity by 125x, and lower latency 200x.
The forum believes those targets are achievable with improved management of the wavelengths used for optical transmissions, increased use of multiplexing, and with multicore fibers – cables that include more optical fibers.
[10]
As explained to The Register by Sean Lawrence, a vice-president at NTT and co-head of its IOWN development office, the IOWN Forum plans four steps.
The first is IOWN powering WANs. NTT already uses IOWN to do this in Japan, using optical transceivers that are either standalone devices or plugins to routers. In either case, signals enter the transceiver as light, and the device transforms it into a familiar and utterly usable electronic signal – Ethernet.
The IOWN all-photonic network is already an open spec so once transceivers become available, carriers can put it to work.
Next comes IOWN as an alternative connector between boards inside computers – essentially a replacement for the PCI bus.
[11]NTT creates a drone that triggers and catches lightning – then keeps flying
[12]NTT Data, IBM team on mainframe cloud for banks
[13]NTT uses scattered monitors to trick your brain into seeing 3D images
[14]Flying phone base stations to take off over Japan in 2026
"We envision an external shared bus instead of an internal PCI bus," Lawrence explained. "Once we get to this point, you don't have to have resources like CPU, GPU, memory in each chassis. We can get to shared pools of resources – and composable servers."
What about the wireless bit?
The "W" in IOWN stands for "wireless," but the tech is mostly about all-photonic networks.
The project includes wireless because its backers know that few organizations will only run wired networks, meaning IOWN's specs need to include details on how it will interface with wireless links and the devices that use them.
If those resources live at the end of an IOWN WAN, Lawrence thinks it may be possible to construct virtual servers with CPUs and memory in different datacenters.
"Whole new compute architectures become possible," Lawrence said, imagining the chance to compose servers as needed by drawing on a pool of processors, GPUs, memory, and storage.
The third phase for IOWN is replacing electronic connections on a motherboard, and the last is optical communication between dies inside a processor.
Getting real this year
A plan to supplant today's networks by 2030 is vastly ambitious. But IOWN is advancing on several fronts.
Lawrence told us he's working to commercialize the IOWN technology that NTT uses to run WANs in Japan, so other carriers around the world can adopt it. He said NTT plans to announce products "in coming months."
IOWN as a PCI replacement is also advancing. Lawrence said NTT is showing a prototype of the tech at Osaka Expo 2025. Broadcom's VMware business unit has worked on pooled memory since it announced " [15]Project Capitola " in 2021 and recently added memory tiering that allows hosts to access capacity on NvME drives in other servers. So phase two of IOWN is also becoming a reality. Phases three and four are more distant.
NTT hopes to play in all four phases because in addition to its well-known tech services businesses and Japan-based carrier operations, it's an electronics manufacturer that wants to commercialize its research by creating components IOWN Forum members can use to create IOWN-ready products.
Many of the forum's 150-plus members would like to do the same.
For Register readers, IOWN may first manifest in networking devices that include or are compatible with transceivers that use the project's all-photonic network spec. And if the IOWN Forum's plans become reality, one day the tech will be everywhere. ®
Get our [16]Tech Resources
[1] https://www.ntt.com/en/about-us/press-releases/news/article/2024/1205.html
[2] https://group.ntt/en/newsrelease/2024/12/09/241209a.html
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[6] https://www.rd.ntt/e/brl/latesttopics/2019/04/latest_topics_201904161802.html
[7] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[8] https://iowngf.org/
[9] https://iowngf.org/itu-secretary-general-and-the-iown-global-forum-president-and-chairperson-meet-to-accelerate-further-collaboration
[10] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[11] https://www.theregister.com/2025/04/23/ntt_lightning_drone/
[12] https://www.theregister.com/2024/09/19/ntt_data_ibm_mainframe/
[13] https://www.theregister.com/2024/06/18/ntt_3d_images_display_illusion/
[14] https://www.theregister.com/2024/06/04/aalto_haps_japan/
[15] https://www.theregister.com/2021/10/11/vmware_project_capitola_explained/
[16] https://whitepapers.theregister.com/
Working with Hitachi Vantara, the company [1]virtualized a pair of storage arrays into a single logical unit that achieved real-time data synchronization. That's a neat trick in the datacenter, but NTT and Hitachi pulled it off despite the two arrays being in different facilities separated by 600 km – the kind of distance that creates lots of latency and makes real-time transfers impossible.
Later in the month, NTT [2]detailed its efforts to shift data at an astounding 455 terabits per second across 1,000 km using ordinary optic fibers.
[3]
The common denominator in both tests was technology called "Innovative Optical and Wireless Network" (IOWN), an all-optical networking stack that NTT hopes will mature in 2030 and expects will reduce power consumption by 100x, improve transmission capacity by 125x, and reduce network latency to 0.5 percent of current levels.
[4]
[5]
IOWN draws on research into optical transistors conducted by NTT boffins, whose work made it into the Nature Photonics journal [6]in 2019 .
Transistors are electronic devices, and signals that pass through electronic systems degrade due to electromagnetic interference. Optical communications, by contrast, are largely immune to electromagnetic interference, and can carry data faster and more reliably over long distances.
[7]
But signals carried on optical networks eventually reach electronic devices – and when that happens, the data flow slows.
Through IOWN, NTT plans to stretch optical networks deeper into computing infrastructure. Instead of data sprinting across an optical WAN before strolling through an electronic router and crawling across copper, NTT wants information to spend more time on optical media and switch through optical transistors, and flow at high speed for longer.
NTT created the [8]IOWN Forum to explore and advance such scenarios, and some of the biggest names in IT have joined it: Google, Microsoft, Cisco, Dell, Qualcomm, Intel, Nvidia, and Broadcom are all aboard. Japanese industrial giants like Mitsubishi have signed up. Telco giants BT and Telefonica are members, as are Ericsson and Nokia. Sony, which knows a thing or two about optics after working on Blu-Ray, is contributing research. The International Telecommunication Union last week [9]agreed "to accelerate further collaboration."
From WAN to motherboard
The IOWN Forum has set what it calls "lighthouse targets" to lower network power consumption by 100x, improve transmission capacity by 125x, and lower latency 200x.
The forum believes those targets are achievable with improved management of the wavelengths used for optical transmissions, increased use of multiplexing, and with multicore fibers – cables that include more optical fibers.
[10]
As explained to The Register by Sean Lawrence, a vice-president at NTT and co-head of its IOWN development office, the IOWN Forum plans four steps.
The first is IOWN powering WANs. NTT already uses IOWN to do this in Japan, using optical transceivers that are either standalone devices or plugins to routers. In either case, signals enter the transceiver as light, and the device transforms it into a familiar and utterly usable electronic signal – Ethernet.
The IOWN all-photonic network is already an open spec so once transceivers become available, carriers can put it to work.
Next comes IOWN as an alternative connector between boards inside computers – essentially a replacement for the PCI bus.
[11]NTT creates a drone that triggers and catches lightning – then keeps flying
[12]NTT Data, IBM team on mainframe cloud for banks
[13]NTT uses scattered monitors to trick your brain into seeing 3D images
[14]Flying phone base stations to take off over Japan in 2026
"We envision an external shared bus instead of an internal PCI bus," Lawrence explained. "Once we get to this point, you don't have to have resources like CPU, GPU, memory in each chassis. We can get to shared pools of resources – and composable servers."
What about the wireless bit?
The "W" in IOWN stands for "wireless," but the tech is mostly about all-photonic networks.
The project includes wireless because its backers know that few organizations will only run wired networks, meaning IOWN's specs need to include details on how it will interface with wireless links and the devices that use them.
If those resources live at the end of an IOWN WAN, Lawrence thinks it may be possible to construct virtual servers with CPUs and memory in different datacenters.
"Whole new compute architectures become possible," Lawrence said, imagining the chance to compose servers as needed by drawing on a pool of processors, GPUs, memory, and storage.
The third phase for IOWN is replacing electronic connections on a motherboard, and the last is optical communication between dies inside a processor.
Getting real this year
A plan to supplant today's networks by 2030 is vastly ambitious. But IOWN is advancing on several fronts.
Lawrence told us he's working to commercialize the IOWN technology that NTT uses to run WANs in Japan, so other carriers around the world can adopt it. He said NTT plans to announce products "in coming months."
IOWN as a PCI replacement is also advancing. Lawrence said NTT is showing a prototype of the tech at Osaka Expo 2025. Broadcom's VMware business unit has worked on pooled memory since it announced " [15]Project Capitola " in 2021 and recently added memory tiering that allows hosts to access capacity on NvME drives in other servers. So phase two of IOWN is also becoming a reality. Phases three and four are more distant.
NTT hopes to play in all four phases because in addition to its well-known tech services businesses and Japan-based carrier operations, it's an electronics manufacturer that wants to commercialize its research by creating components IOWN Forum members can use to create IOWN-ready products.
Many of the forum's 150-plus members would like to do the same.
For Register readers, IOWN may first manifest in networking devices that include or are compatible with transceivers that use the project's all-photonic network spec. And if the IOWN Forum's plans become reality, one day the tech will be everywhere. ®
Get our [16]Tech Resources
[1] https://www.ntt.com/en/about-us/press-releases/news/article/2024/1205.html
[2] https://group.ntt/en/newsrelease/2024/12/09/241209a.html
[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0
[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[5] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[6] https://www.rd.ntt/e/brl/latesttopics/2019/04/latest_topics_201904161802.html
[7] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0
[8] https://iowngf.org/
[9] https://iowngf.org/itu-secretary-general-and-the-iown-global-forum-president-and-chairperson-meet-to-accelerate-further-collaboration
[10] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_specialfeatures/cloudinfrastructuremonth&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aJTNlhQsUo37S8glt1szRQAAAMA&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0
[11] https://www.theregister.com/2025/04/23/ntt_lightning_drone/
[12] https://www.theregister.com/2024/09/19/ntt_data_ibm_mainframe/
[13] https://www.theregister.com/2024/06/18/ntt_3d_images_display_illusion/
[14] https://www.theregister.com/2024/06/04/aalto_haps_japan/
[15] https://www.theregister.com/2021/10/11/vmware_project_capitola_explained/
[16] https://whitepapers.theregister.com/
Doesn't matter to me, I'm not paying for "streaming services"
Grunchy
I'll be jiggered if I'm ever gonna pay any streaming service a single solitary cent. Oh, I'll use your damn optical network at the same price it was last time, or maybe a little more.
But instead of streaming anything, I'll just torrent download the full Bob Burger series and watch it off the Android box.
(Oh, hush. I was gonna borrow it from the library and dub it off anyway.)
interesting technology
Anonymous Coward
The link to their 2019 work seems to have foreshadowed a bright future.
I am curious how the latency between the two data centres 600km apart (> 600 km/c ~ 2 ms => rtt > 4 ms) † was handled in their real time synchronisation tests.
I imagine a fair bit of filthy AI lucre might be thrown in this direction as it would appear to solve some of the current LLM bottlenecks.
† I think the refractive index of glass is ~1.5 so probably > 3 ms and > 6 ms
shared pools of resources – and composable servers
steelpillow
I remember mainframes like that in the late 1960s. Usually strung around in just the one big room, with the codies sat at card or tape punchers next door, and the power supplies through the door on the other side.
SoC is inherently 1000 times faster because the signal paths are 1000 x shorter. Given that the speed of EM waves through copper is over half that of fibre, I am puzzling how this 500 times extra acceleration comes about. Presumably lots of clever stuff they aren't telling us, so maybe us hoi polloi could apply all that to copper and settle for a slightly smaller reach and a vastly smaller bill?
Super
Paris to Rome, Paris to Madrid, Paris to Berlin, Paris to Copenhagen etc ... are all about 1,000 km. Getting 389 Tb/s on 12-core fibers with femto-farad O-E-O optical transistors (nano-PhotoDetectors/PD + nano-Electro-Optic Modulator/EOM) between those major EU cities sounds wonderful to me.
Can't wait for such IOWN all-photonic networking to become reality (rather than some ever-receding hairline future ...)!