计算机科学与技术系

Department of Computer Science and Technology

Education background

Bachelor of Applied Mathematics, Taiyuan Institute of Technology, Taiyuan, China, 1984;

Master of Applied Mathematics, Harbin Institute of Technology, Harbin, China, 1987;

Ph.D. in Computer Architecture, Harbin Institute of Technology, Harbin, China, 1996.

Social service

High-Performance Computing Institute, Department of Computer Science & Technology: Director (2004-);

Academic Committee of Department of Computer Science & Technology: Deputy Director (2008-);

"High-Performance Computer and Core Software" Project, National 863 High-Tech Program: Member of Experts Group (2002-2005);

"High-Productivity Computer and Grid Services Environment" Project, National 863 High-Tech Program: Member of General Experts Group (2006-);

GCC 2008: Program Chair (2008).

Areas of Research Interests/ Research Projects

Parallel and distributed computing

Earth System Model

National Basic Research Program of China (The 973 Program): Semantics-based Information Query and Integration in Information Grid (2004-2009);

National Natural Science Foundation of China: Application Middleware for Grid Computing (2004-2007);

National 863 High-Tech Program: High-Performance Computing Environment for Supporting Earth System Research (2010-2012);

National 863 High-Tech Program: Research on Service Technologies and System Integration of China National Grid (2007-2010);

National 863 High-Tech Program: Research and Development of China National Grid-Data Grid (2007.01-2009.12).

Research Status

I have directed many research projects sponsored by National Basic Research Program, National Natural Science Foundation, and National 863 High-Tech Program since 2003. Based on the above work, we have published more than 85 papers in referred international journals like Journal of Grid Computing, Concurrency and Computation: Practice & Experience, Nucleic Acids Research, and Nano Letters, and conferences like CCGrid, Grid, ICNP, ICDCS, IPTPS, and ICPP. Our key achievements are as follows:

1. We have researched on the ways of analyzing the performance of computing grid and put forward a range-based approach to predicting grid performance. The results were published in Grid and CCGrid. I have also directed the development of Corsair, the first cloud storage product system in China. Now, Corsair is of 100TB storage with more than 15,000 registered users and 300 communities. Each day, it has more than 3,000 visits and as high as 1.3TB data traffic.

2. We have developed a Java-based parallel computing environment that enables users to transparently compile and run parallel programs on heterogeneous clusters. We have studied the ways to analyze and optimize high-performance computing programs. Our joint work with Department of Materials has produced a published paper "Icosahedral FACE-Centered Cubic Fe Nanoparticles" in Nano Letters, which is one of the most famous journals on nano research.

3. I have put forward a new multi-level partitioning (MLP) strategy for facilitating keyword searching in structured peer-to-peer systems. Compared with query flooding and inverted list intersection schemes, MLP can dramatically reduce bisection backbone bandwidth and communication latency. The paper about our detailed work in this area was published in IPTPS 2004, and has got 58 citations so far according to Google Scholar.

4. In earth system simulation, we have optimized the high-resolution atmospheric model gamil (360*180) and made it possible to efficiently utilize 480 CPU cores. In earth system model, we have developed many high-performance interpolation algorithms for couplers. All the above work provides a good basis for earth system research.

Honors And Awards

National Science and Technology Progress Award, Second Class-China National Grid (2007);

National Science and Technology Progress Award, Second Class-High-Performance Cluster and Mass Storage System (2007);

CVIC Software Talent Award (2006);

Tsinghua Outstanding Teaching Award for Young Teachers (2005).

Academic Achievement

[1] Guangwen Yang, Hai Jin, Minglu Li, Nong Xiao, Wei Li, Zhaohui Wu, Yongwei Wu. Grid Computing in China. Journal of Grid Computing, vol. 2, no. 2, pp. 193-206, 2004.

[2] M. Chen, G.W. Yang. Reciprocity: Enforcing contribution in P2P perpendicular downloading. IEICE Transactions On Information And Systems, vol. E89D, no. 2, pp. 563-569, 2006.

[3] B.Y. Zhang, G.W. Yang, W.M. Zheng. Jcluster: an efficient Java parallel environment on a large-scale heterogeneous cluster. Concurrency and Computation: Practice & Experience, vol. 18, no. 12, pp. 1541-1557, 2006.

[4] J.J. Jiang, G.W. Yang. An optimal replication strategy for data grid systems. Frontiers of Computer Science in China, vol. 1, no. 3, pp. 338-348, 2007.

[5] Tao Ling, Lin Xie, Jing Zhu, Huimin Yu, Hengqiang Ye, Rong Yu, Zhiying Cheng, Li Liu, Li Liu, Guangwen Yang, Zhida Cheng, Yujia Wang and Xiuliang Ma. Icosahedral Face-Centered Cubic Fe Nanoparticles: Facile Synthesis and Characterization with Aberration-Corrected TEM. Nano Letters, vol. 9, no. 4, pp. 1572-1576, 2009.

[6] Yongwei Wu, Yulai Yuan, Guangwen Yang, Weimin Zheng. An Adaptive Task-level Fault-tolerant Approach for Grid. Journal of Supercomputing, vol. 51, no. 2, pp. 97-114, 2010.

[7] S.M. Shi, G.W. Yang, D.X. Wang, J. Yu, S.G. Qu, M. Chen. Making peer-to-peer keyword searching feasible using multi-level partitioning. Proc. the 3rd International Workshop on Peer-to-Peer Systems (IPTPS'04), San Diego, CA, USA, 2004, Lecture Notes In Computer Science, vol. 3279, pp. 151-161.

[8] Y.W. Wu, Y.L. Yuan, G.W. Yang, W.M. Zheng. Load Prediction Using Hybrid Model for Computational Grid. Proc. the 8th IEEE/ACM International Conference on Grid Computing (Grid 2007), Austin, Texas, USA, 2007, pp. 235-242.

[9] Yulai Yuan, Yongwei Wu, Guangwen Yang, Weiming Zheng. Adaptive Hybrid Model for Long Term Load Prediction in Computational grid. Proc. IEEE Symposium on Cluster Computing and the Grid (CCGrid 2008), Lyon, France, 2008, pp. 340-347.

[10] Yulai Yuan, Guangwen Yang, Yongwei Wu, Weimin Zheng. PV-EASY: A Strict Fairness Guaranteed and Prediction Enabled Scheduler in Parallel Job Scheduling. Proc. the 19th ACM International Symposium on High Performance Distributed Computing (HPDC 2010), Chicago, Illinois, USA, 2010, to appear