Keynote Speakers

Abstract

Reevaluating Amdahl's Law in the Multicore Era

Multicore architecture has become the trend of high performance processors. While it is generally accepted that we have entered the multicore era, concerns exist on when or will moving into the manycore stage. Recently, Hill and Marty presented a pessimistic view of multicore scalability, citing Amdahl's law and the memory-wall problem. Technology is available, but major vendors are hesitant in making processors that have a large number of cores. This is a very interesting phenomenon, where history seems to repeat itself on the scalability debate of parallel processing that occurred 20 years ago. In this introductory keynote talk we first review the history and concepts of scalable computing, and review the current technologies and the memory-wall problem. We then use the same hardware cost model of multicore chips used by Hill and Marty to introduce two performance models from the scalable computing point of view. These models show that there is no inherent, immovable upper bound on the scalability of multicore architectures. Finally, we conclude with proposed solutions to the memory-wall problem to make the potential scalability of multicore reachable in practice.

Biography

Dr. Xian-He Sun is a professor of Computer Science at the Illinois Institute of Technology (IIT). He received his BS in Mathematics in 1982 from Beijing Normal University, P.R. China, and completed his MS in Mathematics, MS and Ph.D. in Computer Science in 1985, 1987, and 1990, respectively, all from Michigan State University. He was a post-doctoral researcher at the Ames National Laboratory, a staff scientist at the ICASE, NASA Langley Research Center, an ASEE fellow at the US Navy Research Laboratories, and was an associate professor and the founding director of the Scalable Computing Software laboratory in the Department of Computer Science, Louisiana State University before he joined the Computer Science Department, IIT in August 1999. Currently he is a professor at IIT, a guest faculty in the Mathematics and Computer Science Division at the Argonne National Laboratory, a visiting scientist at the Fermi National Accelerator Laboratory, and the director of the Scalable Computing Software laboratory at IIT. His research interests include parallel and distributed processing, middleware, performance evaluation, and high end computing.

Dr. Sun has published over one hundred research articles in the field of computer science and communication, has ten granted and pending US and international patents, and his research is supported by NSF (NSF-JEAP, ACS-9720215, CCR-9972251, EIA-0130673, ANI-0123930, ACI-0130458, EIA-0224377, ACI-0305355, CNS-0406328, SCI-0504291,CNS-0509118, CCF-0621435, CCF-0702737, CNS-0751200, CNS-0834514), and other US government agencies. His research in mobility of legacy code is one of the first nine projects supported by the National Science Foundation under the Middleware Initiative program. His PDD and PPT algorithms have been included in IBM's Parallel Engineering and Scientific Software Library (PESSL) and other commercial and research software packages as a community standard. His memory-bounded (the so-called Sun-Ni's law) and memory access delay performance model are introduced in many modern textbooks as a must known in performance evaluation of scalable computing systems. Chicago Sun-Times called his work in cross-network service as turning "POTS (plain Old Telephone service) into PANS (Pretty Amazing New Stuff), moving (landline) phone into the internet loop" (May 12, 2003). An IETF Internet standard (RFC--RFC3910) was officially released in 2004 based on his novel concept of cross-network service. His current research includes the development of the Data Access Server system for high performance machines, the development of the Fault awareness ENabled Computing Environment (FENCE) for high end computing, and the development of the Dynamic Virtual Machine (DVM) middleware for distributed computing. His group received the ACM/IEEE HPC award and the NSF V-Tech HEC Challenge award in 2007 and 2006 for its contribution in the Data Access Server project and the FENCE project, respectively.

Dr. Sun is a senior member of IEEE, a member of ACM and PHI KAPPA PHI, and was a distinguished speaker of IEEE CS society from 2001 to 2003. He is an editor of five international professional journals, has served and is serving as the chairman or a member of program committee for numerous international conferences and workshops, and served and is serving on many boards and committees of goverment agencies and private fundations for research activities. He received the ONR and ASEE Certificate of Recognition award in 1999, the Best Paper Award at the International Conference on Parallel Processing in 2001, the Best Poster Award at IEEE International SuperComputing Conference in 2003, and the IIT Dean's Excellence Award for Research in 2006. He is a Tan Chin Tuan Fellow, Nanyang Technological University, Singapore, a Guest Professor, University of Science and Technology of China (USTC) and Beijing University of Aeronautics and Astronautic, P. R. China, and the president of the Society of Chinese-American Professors & Scientists.

Abstract

Autonomic Cloud Systems Management: Challenge and Opportunities

Cloud computing, unlocked by virtualization, is emerging as an increasingly important service-oriented computing paradigm. Management is key to providing accurate service availability and performance data and to enabling on-demand real-time capacity planning to meet service demands dynamically. This is because virtualization does not reduce the complexity of a system. In fact, having multiple virtual machines (VMs) running on top of a shared physical computing infrastructure increases the overall system complexity and poses new challenges in systems management. Optimizing one component may compromise the others, leading to overall performance degradation. Frequent component failures here and there would even cause low system productivity.

This talk starts with a review of challenge issues in the management of large scale cloud computing systems. A machine learning approach is introduced for tackling the performance and reliability problems. Two case studies will be presented. One is anomaly detection, bottleneck identification, and VM autoconfiguration. The other is proactive failure management that deals with failures before they occur in cloud systems. Empirical models built from statistical learning exhibit great potential to help overcome the challenges of scale and complexity in current and future networked computer systems.

Biography

Dr. Cheng-Zhong Xu is a Professor in the Department of Electrical and Computer Engineering of Wayne State University, the Director of the Laboratory for Cloud and Internet Computing, and the Director of Sun's Center of Excellence in Open Source Computing and Applications (SUN OSCA). He received his BS and MS degrees from Nanjing University, and Ph.D. from the University of Hong Kong in 1986, 1989, and 1993, respectively, all in Computer Science. He was a Guest Professor of the Paderborn Center of Parallel Computing in the Paderborn Univeristy, Germany before he joined the faculty of Wayne State University in 1995.

Dr. Xu's current research interest includes resource management in distributed and parallel systems, high performance cluster computing, and scalable and secure Internet services. He has published extensively in these areas. He was a main inventor and promoter of diffusive load balancing algorithms in parallel computing, and a key innovator of feedback control and machine learning-based systems management approaches for the provisioning of high service avaiability of Internet servers and datacenters. Dr. Xu was the leading author of the book "Load Balancing in Parallel Computers" (Kluwer Academic/Springer Verlag, 1995). It was arguably the first that addressed the load balancing issue systematically. Dr. Xu recent book "Scalable and Secure Internet Services and Architecture" (Chapman & Hall/CRC Press, 2005) provided an in-depth analysis of the Internet services in a unified framework from the performannce perspective. Dr. Xu's research was supported by US NSF, NASA, and industries like Cray Research and Sun Microsystems. He is the Director of the Sun OSCA Center, established by Sun Microsystems in partnership with Wayne State University.

Dr. Xu is an editor of a number of leading journals in his areas, including IEEE Transactions on Parallel and Distributed Systems and Journal of Parallel and Distributed Computing. He is a recipient of "Faculty Research Award" of Wayne State University in 2000, President's Award for Excellence in Teaching in 2002, and Career Development Chair Award in 2003.

Abstract

Metropolitan VANET: Services on the Road

The application of mobile communication technology to support road traffic constitutes a challenging, but at the same time very promising working area for research and development. Services reach from vehicular safety applications including collision and other safety warnings to non-safety applications like real-time traffic congestion and routing information, high-speed tolling, mobile infotainment, and many others. The creation of high-performance, highly reliable, highly scalable, and secure wireless vehicular ad hoc network (VANET) technologies, though, presents an extraordinary challenge to the wireless research community: a high degree of communication reliability is needed under unfavorable channel conditions. Clearly, the specificity of VANETs in terms of mobility behavior and applications scenarios and requirements makes VANET research an exciting and demanding application- and purpose-driven sub-discipline of wireless networking. In this talk, we will discuss the challenges and opportunities of VANETs in metropolitan scales. A concrete case in Shanghai will also be studied.

Biography

Minglu Li graduated from the School of Electronic Technology, University of Information Engineering, in 1985 and received the PhD degree in computer software from Shanghai Jiao Tong University (SJTU) in 1996. He is a full professor and the vice dean of School of Electronic Information and Electrical Engineering, the director of Grid Computing Center, and the executive director of 973 Wireless Sensor Networks Jiont Lab of SJTU. Currently, his research interests include grid computing, services computing, and wireless sensor networks. He has presided over 20 projects supported by the National Natural Science Foundation, 863 Program, 973 Program, and Science and Technology Commission of Shanghai Municipality (STCSM). He has published more than 100 papers in academic journals and international conferences. He is also a member of the Expert Committee of the ChinaGrid Program of Ministry of Education, a principal scientist of ShanghaiGrid, which is a grand project of STCSM, the Executive Committee of the China Grid Forum, and the Executive Committee of the Technical Committee on Services Computing of the IEEE Computer Society.