By IBT Staff Reporter
Published 09/08/26 AT 11:13 PM JST

Share on FacebookShare on LinkedInShare on RedditShare on FlipboardShare on Pocket

KEY POINTS

• Daisan and ONESTRUCTION begin collaboration on data-center digital twins using NVIDIA Omniverse libraries
• Integrated simulation environment uses NVIDIA DSX Sim Blueprint and Vera Rubin DSX reference design
• System targets real-time evaluation of heat, airflow and PUE in hybrid liquid- and air-cooled data centers

A digital twin simulation of a data center. Note: Japanese text in the original image from press release has been translated into English by our editorial team. Daisan Co.

Daisan Co. and ONESTRUCTION Inc. have begun collaborating to build digital twins for data centers using NVIDIA Omniverse libraries, Daisan said on September 8, as demand for new facilities in Japan expands with the spread of generative artificial intelligence.

The companies plan to develop an integrated simulation environment based on NVIDIA DSX Sim Blueprint and the Vera Rubin DSX reference design. The platform is intended to connect data-center design and simulation in a single 3D environment and support customer review and consensus-building by enabling real-time visualization and assessment of heat, airflow and power usage effectiveness, or PUE, in hybrid liquid- and air-cooled facilities, according to the release.

Daisan, based in Osaka, and Tottori-based ONESTRUCTION said data-center planning and design require evaluation across a wide range of factors, including thermal and airflow management, server-rack layout and power-efficiency optimization. They said those processes have conventionally relied on separate simulations and engineering judgment.

Daisan said it will continue to pursue data-center equipment construction projects. The company also plans to apply digital twins beyond the design stage to construction planning, construction management and operational improvement in the future.

Japan has seen growing investment interest in data centers as AI workloads increase demand for computing infrastructure and power-intensive server capacity. The shift toward liquid cooling has also gained attention as operators prepare facilities for higher-density AI servers.