Practical Performance Confirmed for Remote Distributed AI Infrastructure Between Tokyo and Fukuoka Utilizing 'IOWN APN', Tailored to Workload Characteristics
Four companies—GMO Internet, NTT East, NTT West, and QTnet—have completed a technical demonstration of a remote distributed AI infrastructure between Tokyo and Fukuoka using IOWN APN. They confirmed that for large language model training, performance degradation was limited to approximately 0.5% compared to local environments, proving the feasibility of practical AI development in remote settings.
📋 Article Processing Timeline
- 📰 Published: March 30, 2026 at 19:31
- 🔍 Collected: March 30, 2026 at 22:56 (3h 24m after Published)
- 🤖 AI Analyzed: April 22, 2026 at 22:31 (551h 35m after Collected)
GMO Internet Inc. (Headquarters: Shibuya-ku, Tokyo; Representative Director, President and Executive Officer: Tadashi Ito; hereinafter, GMO Internet) of GMO Internet Group, NTT East Corporation (Headquarters: Shinjuku-ku, Tokyo; Representative Director, President: Naoki Shibuya; hereinafter, NTT East), NTT West Corporation (Headquarters: Osaka-shi, Osaka; Representative Director, President: Ryota Kitamura; hereinafter, NTT West), and QTnet, Inc. (Headquarters: Fukuoka-shi, Fukuoka; Representative Director, President: Yoshio Ogura; hereinafter, QTnet) have completed a technical demonstration of a remote distributed AI infrastructure between Tokyo and Fukuoka utilizing "IOWN (Innovative Optical and Wireless Network)" and its "APN (All-Photonics Network)".
In this demonstration, conducted from November 2025 to February 2026, an IOWN APN actual line was laid between Tokyo (storage) and Fukuoka (GPU). The AI workload performance was measured and evaluated on an AI development platform connecting GMO GPU Cloud's GPUs with large-capacity storage. As a result, for large language model (LLM) training, performance degradation was limited to approximately 0.5% compared to local environments, confirming that the impact was extremely limited. For image classification tasks involving data loading, it was also confirmed that practical level processing is possible even in remote environments through optimization of training data, etc., demonstrating that practical AI development in a remote distributed environment is possible with designs tailored to workload characteristics.
Prior to this demonstration, the four companies conducted a preliminary demonstration (Phase 1) in July 2025, performing performance tests in a simulated remote environment assuming a Tokyo-Fukuoka distance (approx. 1,000 km), and published the details as a technical report.
Press Release:
"Start of Technical Demonstration for Next-Generation Distributed AI Infrastructure Utilizing 'GMO GPU Cloud' and Low-Latency Line 'IOWN APN'"
Technical Report:
"Details and Results of GPU-Storage Interconnection Performance Test in a Simulated Remote Environment Utilizing 'IOWN APN'"
Based on the results of this demonstration, the four companies will continue to work towards the practical application of remote distributed AI infrastructure tailored to customer needs.
## [Background and Purpose]
With the recent spread of generative AI and large language models (LLMs), the demand for AI development platforms is rapidly expanding. Traditionally, GPUs and large-capacity storage have been required to be physically co-located. However, to address data center space constraints and the need for companies to manage data at their own sites, there is a demand for distributed AI development platforms that transcend geographical limitations. The four companies aim to realize this with IOWN, which features high speed, large capacity, and low latency.
In this demonstration, conducted from November 2025 to February 2026, an IOWN APN actual line was laid between Tokyo (storage) and Fukuoka (GPU). The AI workload performance was measured and evaluated on an AI development platform connecting GMO GPU Cloud's GPUs with large-capacity storage. As a result, for large language model (LLM) training, performance degradation was limited to approximately 0.5% compared to local environments, confirming that the impact was extremely limited. For image classification tasks involving data loading, it was also confirmed that practical level processing is possible even in remote environments through optimization of training data, etc., demonstrating that practical AI development in a remote distributed environment is possible with designs tailored to workload characteristics.
Prior to this demonstration, the four companies conducted a preliminary demonstration (Phase 1) in July 2025, performing performance tests in a simulated remote environment assuming a Tokyo-Fukuoka distance (approx. 1,000 km), and published the details as a technical report.
Press Release:
"Start of Technical Demonstration for Next-Generation Distributed AI Infrastructure Utilizing 'GMO GPU Cloud' and Low-Latency Line 'IOWN APN'"
Technical Report:
"Details and Results of GPU-Storage Interconnection Performance Test in a Simulated Remote Environment Utilizing 'IOWN APN'"
Based on the results of this demonstration, the four companies will continue to work towards the practical application of remote distributed AI infrastructure tailored to customer needs.
## [Background and Purpose]
With the recent spread of generative AI and large language models (LLMs), the demand for AI development platforms is rapidly expanding. Traditionally, GPUs and large-capacity storage have been required to be physically co-located. However, to address data center space constraints and the need for companies to manage data at their own sites, there is a demand for distributed AI development platforms that transcend geographical limitations. The four companies aim to realize this with IOWN, which features high speed, large capacity, and low latency.