Press Release
The Open Compute Project (OCP) Global Summit will return from October 14 to 16 at the San Jose Convention Center, gathering worldwide innovators to discuss the latest breakthroughs in open computing, energy efficiency, storage, networking, and security. This global platform highlights how open collaboration accelerates technological transformation and sets the stage for the future of computing.
As a leader in AI server technology, Inventec is dedicated to pushing the boundaries of innovation and delivering solutions that empower customers in the age of artificial intelligence. With the exponential growth of AI driving unprecedented demand for servers, rack systems, and cooling technologies, Inventec will join this year’s OCP Global Summit as a Diamond Sponsor, showcasing a full portfolio of next-generation rack solutions, AI servers, and thermal innovations.
One of the centerpiece demonstrations at the Inventec booth will be the Artemis II rack solution. Powered by the NVIDIA GB300 NVL72 system and built on NVIDIA MGX platform, Artemis II is purpose-built to address the extreme computational intensity of large-scale generative AI training and inference workloads. Together with the P9000-series rack solution, which integrates AMD EPYC™ 9005 Series and Intel® Xeon® 6 processors with NVIDIA Blackwell and Blackwell Ultra GPUs, these offerings deliver versatile, scalable infrastructure designed for cloud-scale data centers, advanced AI factories, and enterprise-level AI deployments. These solutions empower organizations to accelerate innovation while maintaining performance and efficiency at massive scale.
In addition, Inventec will showcase a proof-of-concept cluster featuring Inventec Edge systems accelerated by the NVIDIA IGX Orin platform in a standard rack. This cluster provides an industrial-grade AI platform for edge computing with the compute power, safety, and security required for robotics workload. With real-time decision-making, safe human-machine collaboration, and Holoscan-enabled responsive sensing, the cluster is designed for easy scalability by adding IGX systems. Seamless integration with Inventec P90G6-SKU2 – an NVIDIA OVX system — bridges edge operations with AI training and simulation. At its core, the Inventec EdgePro management node, built on Intel Xeon 6 processors and compliant with OCP DC-MHS specifications, coordinates workloads with modular design, high compute density, and energy efficiency. Together, this physical AI framework combines OVX’s digital twin simulations with IGX’s enterprise-ready edge deployments, setting a new benchmark for building and scaling robotic intelligence.
Beyond rack-level solutions, Inventec will also introduce three advanced servers. The P4200G7, a 4U RTX PRO Serverbased on the NVIDIA MGX architecture, leverages the NVIDIA RTX PRO 6000 Blackwell Server Edition GPU , NVIDIA ConnectX-8 SuperNIC switch and Intel® Xeon® 6 processors, supporting versatile GPU configurations for enterprise and research applications. The K905G6, a dense multi-server platform, delivers exceptional performance in a compact 2U chassis, supporting AMD EPYC™ 4th and 5th Gen CPU, and up to four half-wide 1U 2P nodes or two half-wide 2U 2P nodes. With flexible power and cooling, including Direct Liquid Cooling, it maximizes efficiency while lowering the total cost of ownership. Rounding out the lineup, the P8500G6 AI server supports AMD Instinct MI300 to MI350 Series GPUs, offering flexible, scalable performance to meet the rising demands of modern AI workloads.
Visit Inventec at booth B50 to explore these cutting-edge technologies firsthand and engage with our team of experts. We look forward to showcasing how Inventec is driving the future of AI servers and cooling solutions at this year’s OCP Global Summit.
