“Data centers accounted for around 1.5% of the world’s electricity consumption in 2024, or 415 terawatt-hours (TWh)... Globally, data center electricity consumption has grown by around 12% per year since 2017, more than four times faster than the rate of total electricity consumption.”

— International Energy Agency, Energy and AI Report, April 2025

Data centers keep the modern world running, but their constant appetite for electricity makes them one of the hardest parts of the digital economy to decarbonize. Even as renewable energy expands, much of it goes unused when supply and demand fall out of step.

Sunlight, wind, and grid capacity change from hour to hour, yet most computing systems still operate in fixed locations. It makes for a frustrating quandary: clean energy is available, but the machines that could use it may be hundreds of kilometers away.

NTT, NTT West, and QTnet are determined to change that. In a recent field test, the three organizations connected data centers in Fukuoka and Osaka, Japan, using the IOWN All-Photonics Network (APN) and demonstrated that workloads can move automatically between them. Every thirty minutes, a specially designed algorithm monitored renewable-power levels and shifted virtual machines and applications to whichever site could run them most efficiently.

IOWN + Algorithm = 30% Increase in Renewable Energy Use

And the result? Without stopping services or lengthening computing time, the system achieved about a 30% rise in renewable-energy use compared with a fixed setup.

How it Works

But we're getting ahead of ourselves, so let's start at the beginning. The project began with a simple question: if the makeup of energy is constantly changing, why can’t data follow it?

Conventional data centers are tied to local grids. When solar or wind output falls, they switch to thermal power and keep running. At the same time, other regions may be generating excess green electricity that could be used. The test team wanted to see if data center operations could respond to these shifts in real time and keep workloads as close as possible to the cleanest available power.

What makes that possible? The fundamental enabler is the IOWN APN, a photonics-based network that transmits information entirely as light from end to end. Its speed and stability allow servers in distant cities to behave as though they were part of one facility.

On top of the APN, NTT developed energy‑cooperative ICT resource control technology that continuously checks each site’s renewable output, electricity prices, and server load, then recalculates where every task should run.

Connecting Fukuoka and Osaka

In the demonstration, virtualized servers in Fukuoka, on the island of Kyushu, and Osaka, on the Japanese mainland, were linked across roughly six hundred kilometers of optical fiber. Each partner played an important role: NTT West built and evaluated the inter‑data‑center network and server environments using the APN; NTT developed and tested the software and workload control system; and QTnet provided the optical fiber link and the Fukuoka‑side data center environment.

The control software compared real data from a day when Kyushu’s renewable generation had been curtailed and redistributed computing accordingly. Within two minutes, it was able to create a full-day operation plan, relocating workloads in thirty-minute cycles. The software proved that live applications, including AI inference tasks, could migrate smoothly between regions without interruption.

The experiment also showed how the method might be scaled. By adding more data centers, workloads could concentrate in areas with surplus renewable energy, or move away from regions facing blackouts or high grid prices. In emergencies, meanwhile, processing could shift automatically to unaffected sites, keeping services online while helping balance the national grid.

Why this Matters

As AI models grow and data processing expands, electricity demand from computing continues to climb. The ability to treat multiple data centers as one flexible resource could potentially reduce both emissions and operational waste: instead of building new facilities or over-provisioning capacity, operators would have the ability to use existing sites more intelligently, matching them with the ebb and flow of clean energy.

Think of a commuter system that dispatches trains from whichever station has the most sunlight powering its lines. The schedule changes constantly, but the passengers still arrive on time. In much the same way, the IOWN APN lets computing tasks follow the energy that powers them.

Looking Ahead

The demonstration matches well with Japan’s wider effort to balance regional energy supply and demand, while at the same time pushing for carbon neutrality. It also reflects NTT’s long-term IOWN vision—integrating photonics, computing, and communication into a single system that works in harmony with natural resources.

As data use grows, so does the need for solutions that are not only powerful but responsible. By letting information travel to where energy is cleanest, IOWN technology shows us a digital infrastructure that adapts as fluidly as nature itself. The goal is not simply to run faster, but to run smarter. Allowing the networks driving human progress to work with the lightest possible footprint.

Innovating a Sustainable Future for People and Planet

IEA: Energy and AI — Executive Summary
https://www.iea.org/reports/energy-and-ai/executive-summary

For further information, please see this link:
https://group.ntt/en/newsrelease/2025/06/11/250611a.html

If you have any questions on the content of this article, please contact:
NTT WEST Corporation
Public Relations
nttw-press@west.ntt.co.jp

NTT Information Network Laboratory Group
Public Relations
https://tools.group.ntt/en/news/contact/index.php

Daniel O'Connor

Daniel O'Connor joined the NTT Group in 1999 when he began work as the Public Relations Manager of NTT Europe. While in London, he liaised with the local press, created the company's intranet site, wrote technical copy for industry magazines and managed exhibition stands from initial design to finished displays.

Later seconded to the headquarters of NTT Communications in Tokyo, he contributed to the company's first-ever winning of global telecoms awards and the digitalisation of internal company information exchange.

Since 2015 Daniel has created content for the Group's Global Leadership Institute, the One NTT Network and is currently working with NTT R&D teams to grow public understanding of the cutting-edge research undertaken by the NTT Group.