Confirmation of Practical Performance According to Workload Characteristics in Remote Distributed AI Infrastructure Demonstration Between Tokyo and Fukuoka Utilizing 'IOWN APN'
NQ Score
42/100
N1 Content Completeness
4
AI Summary (NQ-processed)
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 (LLM) training, performance degradation was limited to approximately 0.5% compared to local environments, and practical-level AI development is possible in a remote distributed environment.
AI Analysis
Frequently Asked Questions
- Q: What is IOWN APN?
- A: IOWN APN is a high-speed, large-capacity, and low-latency all-photonic network technology proposed by NTT. It maximizes the use of optical technology to integrate information processing and communication.
- Q: What was confirmed in this demonstration?
- A: It was confirmed that using IOWN APN to connect remote GPUs and storage between Tokyo and Fukuoka results in extremely limited performance degradation for AI workloads, such as LLM training.
- Q: What are the benefits of this technology?
- A: This technology eliminates the physical constraints of data centers, enabling the construction of high-performance AI development environments in geographically distant locations. This improves the flexibility and efficiency of AI development.