Compal is showing a new NVIDIA Rubin NVL8 system at the OCP Global Summit as part of a broader push to shorten the time between new AI silicon and deployable infrastructure.
In an October 14 announcement, Compal said its OG231-2-L1 combines Intel processors with NVIDIA HGX Rubin NVL8 in a 2U liquid-cooled server. The company is also showing a 44U Rubin NVL8 rack, a 1U NVIDIA Vera CPU tray, and an NVIDIA Spectrum-X SN6600 switch with up to 128 ports of 800GbE connectivity.
NVIDIA’s GTC Taipei keynote shows Rubin systems and rack-scale infrastructure entering production.
NVIDIA’s HGX specifications list Rubin NVL8 with eight Rubin GPUs, 2.3 TB of HBM4, 176 TB/s of memory bandwidth, sixth-generation NVLink, and up to 400 PFLOPS of NVFP4 inference performance. Compal’s message at OCP is that the integration work around those parts—cooling, power, networking, validation, and rack design—is becoming as important as the accelerator itself.
The showcase lands in the same infrastructure cycle as rack-level liquid cooling for AMD Helios, 102.4T AI switching, and 12.8 Tb/s optical links. Compal has not disclosed pricing or a broad availability date for the OG231-2-L1 in today’s release.
BitcoinVersus.Tech Editor’s Note:
We volunteer daily to ensure the credibility of the information on this platform is Verifiably True. If you would like to support to help further secure the integrity of our research initiatives, please donate using the QR code provided
For best QR-code scanning results, please scan this donation code from a desktop or larger screen. Mobile screens may prevent camera recognition.
BitcoinVersus.Tech provides technical education and analysis for informational purposes only.
Leave a Reply