Abstract: Graph computing is a computational model specifically designed for processing graph-structured data, and it is widely applied in fields such as social networks, traffic optimization, and the military. This paper provides an overview of the basic concepts and core tasks of graph computing and reviews its development history. It focuses on the "Tianhe Graph Computing System" (TianheGraph) developed by the National University of Defense Technology. Through key technologies such as hardware-software collaborative graph distribution (GraphCube), topology-aware communication MST, and super graph storage (SuperCSR), TianheGraph significantly improves the efficiency and performance of processing large-scale graph data. The Tianhe graph computing system has won the championship multiple times in the Graph500 rankings, empowering various industries and demonstrating the leading position of domestic supercomputers in the field of graph computing. In the future, graph computing will be integrated with artificial intelligence to drive broader application innovation.