Powering the Future: Why Engineers Are Switching to Ceramic Substrates
Beyond Design: Inductance—The Silent Architect Driving SI and PI
Built for What’s Next: The End of Flat Thinking
Wistron Corporation and Wiwynn Corporation are participating in the 2026 OCP Global Summit in San Jose, California, under the theme “Create AI Beyond Limits,” showcasing a range of AI solutions and technology applications.
The showcase focuses on “Powering the AI Factory: Next-Generation AI Compute and Storage Systems,” featuring next-generation AI compute and storage systems, including:
- NVIDIA Vera Rubin NVL72 Platform
Wistron and Wiwynn are among the first in line with readiness for NVIDIA’s latest generation rack-scale platform, unifying 72 NVIDIA Rubin GPUs, 36 NVIDIA Vera CPUs with NVIDIA NVLink, NVIDIA ConnectX-9 SuperNICs, and NVIDIA BlueField-4 DPU. Built for the resiliency, energy efficiency, and scalability AI factories demand, the fully liquid-cooled platform is optimized for frontier AI model training, inference, and reasoning, delivering breakthrough performance and efficiency for AI factories.
- NVIDIA Vera CPU Rack: Built for Agentic AI and Reinforcement Learning at Scale
NVIDIA Vera CPU rack delivers the CPU infrastructure required to scale reinforcement learning and agentic AI systems. Built on the NVIDIA MGX architecture and integrated with NVIDIA Spectrum-X Ethernet networking, it supports over 22,500 concurrent execution environments per rack—completing workloads twice as fast with twice the efficiency of traditional CPU infrastructure. To secure these environments, NVIDIA Open Agent Safety Platform combines OpenShell runtime controls on Vera CPUs with NVIDIA Sentry on BlueField-4 DPUs for host-independent monitoring and policy enforcement.
- NVIDIA CMX Context Memory Storage Platform
NVIDIA CMX context memory storage platform is a new class of AI-native storage infrastructure designed to accelerate agentic reasoning at scale. Powered by NVIDIA BlueField-4 STX Storage Processor, NVIDIA DOCA, and NVIDIA AI software, NVIDIA CMX extends effective GPU KV cache capacity, enables high-bandwidth sharing of context across the AI pod, and reclaims essential compute and electrical power to support agentic AI inference.
- NVIDIA Spectrum-X Ethernet Photonics
NVIDIA Spectrum-X Ethernet Photonics, based on Spectrum-6, integrates 200G co-packaged optics (CPO) to support 800 GbE networking and was designed for the NVIDIA Rubin generation of NVIDIA’s accelerated computing platforms. NVIDIA Spectrum-X Ethernet networking platforms address congestion at scale, improve reliability through co-packaged optics integration, use direct liquid cooling to increase data center efficiency, and provide the advanced capabilities required for peak AI network performance.
- NVIDIA HGX Rubin NVL8 Platform
NVIDIA HGX Rubin NVL8 platform scales AI factories for enterprise - powered by the extreme co-design of NVIDIA Rubin GPUs, NVIDIA NVLink 6 switch, NVIDIA ConnectX-9 SuperNICs, NVIDIA BlueField-4 DPUs, and NVIDIA Spectrum-X Ethernet - to deliver fully optimized AI and high-performance computing (HPC) application performance with the fastest time to insights.
Wistron cordially invites global industry partners to visit the 2026 OCP Global Summit and explore our latest AI solutions and technology applications, while exchanging insights on the development of next-generation AI infrastructure.
