The 2025 Open Science Computing Alliance Annual Academic Conference was grandly held in Zhengzhou from November 13th to 16th. SWINAI was invited to participate in this academic feast, gaining cutting-edge insights and deepening industry collaboration through the exchanges.
During the conference, the SWINAI team visited the School of Cyberspace Security at Zhengzhou University. They conducted a close-up visit to the school's technological reserves in various fields such as embodied intelligence, intelligent manufacturing, edge visual gateways, and AI-driven inspection robots, along with some already implemented application scenarios. This provided a direct experience of the deep integration between academic research and industrial practice.
The SWINAI team engaged in multi-dimensional, in-depth discussions with several experts. These included Professor Wu Weiguo from Xi'an Jiaotong University, who has profound expertise in supercomputing; Director Cao and Professor Yang from the School of Cyberspace Security at Zhengzhou University; and multiple technical experts from the 32nd Research Institute of China Electronics Technology Group Corporation (CETC). This academic conference focused on core topics such as hyper-intelligence convergence and heterogeneous scheduling. Participants exchanged ideas on constructing software application systems for supercomputing and intelligent computing, discussed the progress and existing challenges in unified management of CPUs and various accelerator cards, as well as heterogeneous GPU virtualization. They also shared valuable experiences and perspectives on issues including the transmission efficiency, consistency, and high availability of distributed data storage, and the practical implementation of visual solutions in specific scenarios.
This participation proved highly rewarding for SWINAI, broadening its technical horizons and establishing bridges for deeper collaboration. SWINAI will continue to collaborate with industry forces and dedicate itself to innovation in the intelligent computing field.
