ASUS JAPAN Plans First Japanese Showing of Fully Liquid-Cooled Rubin NVL8 Server
Редакция ЦОД:Главное
ASUS JAPAN plans to exhibit its XA NR1I-E12R, a fully liquid-cooled 2U AI server based on NVIDIA HGX Rubin NVL8, at CTC DISCOVER 2026 in Japan. The company says availability of the model is planned from autumn 2026.
- ASUS JAPAN plans to demonstrate the fully liquid-cooled 2U XA NR1I-E12R with NVIDIA HGX Rubin NVL8 for the first time in Japan.
- Availability of the XA NR1I-E12R is planned from autumn 2026.
- The event lineup also includes an air-cooled NVIDIA HGX B300 system and an NVIDIA MGX-based GPU server.
- The announcement highlights liquid cooling as a design consideration for high-density GPU deployments.
ASUS JAPAN is scheduled to make its first Japanese showing of the XA NR1I-E12R at CTC DISCOVER 2026, to be held on 16 October at the Grand Prince Hotel Shin Takanawa. The system is a 2U server with 100% liquid cooling and NVIDIA HGX Rubin NVL8.
The company plans to make the XA NR1I-E12R available from autumn 2026. The announcement concerns an exhibition demonstration and planned availability, rather than confirmed commercial shipments, customer deployments or operating installations.
ASUS JAPAN also plans to display the air-cooled XA NB3I-E12 with NVIDIA HGX B300, alongside the ESC8000A-E13P GPU server based on NVIDIA MGX architecture. Among the announced exhibits is a model supporting up to eight Blackwell GPUs, each with a 600W TDP.
The vendor positions full liquid cooling as a means of improving power-use efficiency and containing total cost of ownership as data-center power consumption and heat loads rise. That proposition is increasingly relevant for AI environments: as projects move from proof-of-concept work toward production use, infrastructure requirements could increasingly center on integrated rack and facility readiness for power delivery, thermal management and liquid cooling, rather than compute-platform selection alone.
The planned demonstration reflects the increasing role of liquid cooling in high-density AI infrastructure, where rising compute density and thermal loads are becoming data-center constraints. For operators and infrastructure suppliers, the shift could move procurement decisions beyond GPU platforms alone toward rack-level power delivery, heat rejection and liquid-cooling readiness.
- — / 16 сентября 2026 г. · ЯпонскийПервоисточник →
Материал подготовлен редакцией ЦОД: Главное на основе указанных первоисточников.