The University of Osaka, D3 Center, SANKEN, and NEC Begin Deployment and Pilot Operation of “RED-ONION”

Tokyo & Osaka, August 21, 2026 — The University of Osaka, D3 Center (*1), SANKEN (*2), and NEC Corporation (*3, NEC), have jointly developed RED-ONION (Research EnhanceD Osaka university Next-generation Infrastructure for Open research and innovatioN), a next-generation data transfer infrastructure designed to securely and rapidly transfer large volumes of research data—including sensitive data related to genomics and healthcare, as well as the results of its analysis and processing—in both directions between the D3 Center and SANKEN on the
university’s Suita Campus. The infrastructure is scheduled to begin pilot operation in October 2026.

The infrastructure immediately transfers sensitive data generated at SANKEN to an on-campus data center via a dedicated optical-fiber network with 100 Gbps-class speeds, where the data is stored and managed under robust information security measures. In addition, the infrastructure provides secure connections to high-performance computing systems, including the SQUID and OCTOPUS supercomputers installed at the D3 Center, as well as the mdx II cloud computing platform. This enables highly efficient AI processing and simulation analysis of the data generated on these systems. The resulting processed and analyzed data can also be instantly transferred back to SANKEN. In this way, RED-ONION provides a research and development infrastructure for AI for Science research involving sensitive data.

This data transfer infrastructure will help accelerate research initiatives at SANKEN, including the Nano-Gap Biomolecule Sequencer project led by Professor Masateru Taniguchi of the Department of Bio-Nanotechnology and his colleagues. The project was selected for the Cabinet Office’s Key and Advanced Technology R&D through Cross Community Collaboration Program (K Program) (*4). The research is focused on developing innovative technologies for the highly accurate and rapid analysis of biomolecules. By applying this infrastructure, large volumes of data generated by scientific measurement equipment and research activities can be transferred rapidly and securely, enabling researchers to routinely perform data analysis and processing on high-performance computing systems and develop AI-powered services. This seamlessly connects the processes of research data generation, aggregation, and analysis, significantly improving the productivity of academic research.

In recent years, the volume of data handled in research activities has increased dramatically, driven not only by advances in experimental and measurement equipment, but also by the development of AI and high-performance computing technologies. In particular, in the life sciences and medical fields, large volumes of research data are generated every day through experiments and advanced measurement equipment, much of which includes sensitive data requiring strict management. At the same time, analyzing and utilizing large volumes of research data necessitates connectivity and integration with high-performance computing systems capable of executing computational processing rapidly and efficiently. However, at many universities and research institutions, experimental facilities that generate research data (data sources) and high-performance computing systems, such as supercomputers used for AI processing and simulation analysis computing infrastructure, are geographically and organizationally dispersed. This is due to differences in the specialized expertise, advanced technologies, and physical environments required to develop and operate these systems. Consequently, establishing an environment that seamlessly connects research institutions generating research data with computing centers equipped with high-performance computing systems, enabling the rapid use of large volumes of research data—including sensitive data—for data analysis, large-scale simulations, and AI service development on high-performance computing systems is an urgent priority for accelerating AI for Science research in Japan.

Building on technology research into ultra-high-speed data transfer led by Professor Susumu Date of the D3 Center (Advanced High-Performance Computing Infrastructure System Research Division / Joint Research Laboratory for Integrated Infrastructure of High Performance Computing and Data Analysis) and Associate Professor Keichi Takahashi of the Advanced High-Performance Computing Infrastructure System Research Division, the D3 Center and NEC have collaborated to develop RED-ONION, a next-generation data transfer infrastructure designed to efficiently transfer large volumes of research data. Through repeated demonstration tests and improvements, the platform has now reached the stage of commencing pilot operation at the research facilities of the University of Osaka. NEC has combined the technologies and expertise it has developed over many years with the results of this collaboration to create an ultra-high-speed data transfer solution, the NEC Ultra-High-Speed Data Transfer System (NEC UDTS). NEC aims to put this technology into practical use and deploy it at other universities and research institutions.

This infrastructure connects the D3 Center and SANKEN via a dedicated optical-fiber network, making it possible to transfer large volumes of data between disks by maximizing network performance. When using conventional data transfer methods, data processing on both the source and destination computing servers may not match the transfer speed, even when sufficient network bandwidth is available. This mismatch can hinder the optimal use of bandwidth. In contrast, RED-ONION uses a dedicated engine to centrally control data processing within networks and servers, dramatically accelerating disk-to-disk data transfer between sites by concurrently executing multiple processes. Based on domestic and international verification test results, RED-ONION is expected to enable the transfer of 1 TB of research data in approximately 90 seconds using dedicated networks with 100 Gbps-class speeds.

Beginning in October 2026, the three parties will initiate a pilot operation of this infrastructure. Their goal is to develop a research infrastructure that allows for the secure and flexible transfer and use of sensitive data. This project aims to pave the way for future expansion to other departments and research institutions. Details regarding the start of full-scale operations, target data types, destination computing resources, and operational scope will be finalized sequentially based on pilot operation results. In this way, the D3 Center, SANKEN, and NEC are working to accelerate the research process, lessen the operational burdens on research systems, and effectively use research resources. They are also focused on developing a new model for a research data infrastructure that fulfills the needs of the era of AI for Science. Furthermore, using the insights gained from this development project, the three parties will continue their efforts to enhance Japan’s research capabilities and promote AI for Science. They will also consider opportunities for expansion into life sciences and other research areas involving sensitive data.

Under its value creation model, “BluStellar” (*5), NEC is encouraging the use of AI in scientific research with its BluStellar Scenario aimed at improving research capabilities through the advancement of research information infrastructure. Through these efforts, NEC will facilitate the secure use of research data in academics and support the advancement of research activities.


(*1) The University of Osaka, D3 Center Location: Ibaraki City, Osaka Prefecture; Director: Hideyuki Shimonishi; URL: https://www.d3c.osaka-u.ac.jp/en/
(*2) SANKEN Location: Ibaraki City, Osaka Prefecture; Director: Takeharu Nagai; URL: https://www.sanken.osaka-u.ac.jp/en/
(*3) NEC Corporation Headquarters: Minato Ward, Tokyo; President and CEO: Takayuki Morita; URL: https://group.nec/global/en/
(*4) Cabinet Office, Key and Advanced Technology R&D through Cross Community Collaboration Program (K Program); URL: https://www.jst.go.jp/k-program/program/bio1.html (in Japanese only)

(*5)
"BluStellar" is a value creation model that leverages NEC's cutting-edge technology, honed through years of development and operation and backed by proven, cross-industry expertise, to transform business models. It guides customers toward solving societal challenges and their management issues, leading them into the future.
(Related Site) External: https://group.nec/jp/en/solutions/nec-blustellar/