Steve Chen, a member of the United States National Academy of Engineering, attended the Zhongguancun Forum 2024 and shared his expectations on the development of intelligent computing.
Talking about the impact of large model technology on supercomputers, Chen indicated that large models will promote the development of supercomputers because large models need more and more computing power. However, computing power and capacity will not be enough, intelligent computing will also be needed.
In this regard, neither the United States nor China has yet done so. Chen pointed out that they are heading towards intelligent and distributed systems. “We can start small, perhaps with a city, a community, or a village, as an experiment. Once we succeed, we can connect and parallelize these systems. And we can gain valuable insights.” He hopes China will be the first to invent a distributed intelligent supercomputer.
