Recently, CAS Computer Network Information Center held a special exchange meeting with UCloud Technology Co., Ltd. on AI computing power innovation, computing network optimization, and satellite internet development. The meeting focused on core topics including large-model computing deployment, cross-domain data transmission, and integrated space-ground computing network scheduling, facilitating technical exchanges and discussions on collaboration. 

At the meeting, Ji Xinhua, Chairman and CEO of UCloud, introduced the company's overall development progress, as well as its technological accumulation and strategic plans in the field of AI computing. He shared practical achievements such as distributed inference, end-to-end performance monitoring, PD separation optimization, and rapid model switching, and outlined future visions for building computing centers, empowering AI infrastructure, and deploying space-based computing capabilities. Zhou Xu, Director of the Department of Advanced Network Technologies and Applications Development at the Center, presented research outcomes and practical experiences from the Center’s work in computing network technologies, long-distance RDMA low-latency transmission, distributed caching, and co-optimization experiments between computing and power. 

Through in-depth discussions, both parties reached consensus on leveraging the strengths of industry-academia-research collaboration to advance further technical research and practical implementation. Moving forward, they will deepen cooperation in areas such as distributed inference optimization, cross-network RDMA enhancement, CDN-based inference systems, and satellite internet data transmission, providing critical technical support for the construction of China’s new-generation computing infrastructure and the high-quality development of the artificial intelligence industry.