摘要
Game-tree search plays an important role in the field of Artificial Intelligence （AI）. In this paper, we characterize one parallel game-tree search workload in chess： the latest version of Crafty, a state of art program, on two Intel Xeon shared-memory multiprocessor systems. Our analysis shows that Crafty is latency-sensitive and the hash-table and dynamic tree splitting used in Crafty cause large scalability penalties. They consume 35%-50% of the running time on the 4-way system. Furthermore, Crafty is not bandwidth-limited.
Game-tree search plays an important role in the field of Artificial Intelligence （AI）. In this paper, we characterize one parallel game-tree search workload in chess： the latest version of Crafty, a state of art program, on two Intel Xeon shared-memory multiprocessor systems. Our analysis shows that Crafty is latency-sensitive and the hash-table and dynamic tree splitting used in Crafty cause large scalability penalties. They consume 35%-50% of the running time on the 4-way system. Furthermore, Crafty is not bandwidth-limited.
作者简介
TAN Ying （谭膺） was bom in Sichuan Province, China, in 1981. From 1999, he has been studying at the University of Electronic Science and Technology of China （UESTC） as a graduate student. From March 2005 to Oct. 2005, he joined the Intel China Research Center Micro-Processor Technology Lab as an intern. His research activities have been concerned with computer architecture, program optimization and parallelization, and data mining techniques.LUO Ke-lu （罗克露） is a professor of School of Computer Science and Engineering, UESTC. She graduated in School of Computer Science of Chengdu College of Electronic Engineering in 1982 and was sent to Rome University of Italy by Chinese government for the advanced studies on computer technology in 1985. Her research interests including Real time system, CAD, etc.CHEN Yu-rong （陈玉荣） is a researcher at Microprocessor Technology Lab, Beijing. Currently he conducts research on emerging computing paradigms, parallel algorithms, and scalable workload development and analysis. He joined Intel in 2004. Before that he spent two years doing postdoctoral research work in computer science in Institute of Software, Chinese Academy of Sciences. He received his B.S. degree in Applied Mathematics in 1998 and M.S. and Ph.D. degrees in computational mathematics in 2002, all from Tsinghua University, China.ZHANG Yi-min （张益民） is researcher in Microprocessor Technology Lab, Beijing. He joined Intel in 2000. He received his B.A. degree from Fudan University in 1993, his M.S. degree from Shanghai Maritime University in 1996, and his Ph.D .degree from Shanghai Jiaotong University in 1999, all in Computer Science. He leads a team of researchers working on various statistical computing techniques and their scalability analysis.
