Performance Evaluation of Scalable Systems:

What we know and what we do not know


Xian-He Sun

Department of Computer Science

Illinois Institute of Technology

Chicago, IL 60616




Although rapid advances in grid computing are bringing true scalable systems within grasp, developing scalable software seems to be more elusive now than that of ten years ago. The most significant question of computing is the same today as it has been for many decades: How can software application take advantage of hardware computing power. In this talk, we will study the issues of performance evaluation, measurement, and optimization of scalable system. Existing metrics and methods will be summarized, and new challenges will be discussed. Our recent results in solving two such new challenges will be introduced. They are the memory-LogP model for performance evaluation of advanced hierarchical memory system and the GHS system for performance measurement and prediction in an enterprise grid-computing environment. Experimental results will be presented to illustrate what we know and what we do not know in the area of scalable performance and what are the implications of current findings.




Professor Xian-He Sun receives his Ph.D. degrees in computer science from Michigan State University in 1990. He was a staff scientist at ICASE, NASA Langley Research Center and was an associate professor in the Computer Science Department at Louisiana State University (LSU). Currently he is a professor of computer science and the director of the Scalable Computing Software (SCS) Laboratory at Illinois Institute of Technology (IIT), and is a guest faculty in the Mathematics and Computer Science Division at the Argonne National Laboratory. Dr. Sun's research interests include parallel and distributed processing, software system, performance evaluation, and scientific computing. He has published intensively in the field, and his research is currently supported by DoD, DoE, NSF, and other government agencies. He has served and is serving as the chairman or on the program committee for a number of international conferences and workshops. He received the ONR and ASEE Certificate of Recognition award in 1999 and the Best Paper Award from the International Conference on Parallel Processing (ICPP01) in 2001.