Keynote Speeches:

   
 

Prof. Xiaodong Zhang, Ohio State University, USA                 

   ---- "Balancing System Resource Supply and Demand for Cost Effective High Performance Computing"
 

Abstract of the Speech

Biographical details of the Speaker

 

The Slides of the Speech

 

Prof. Francis Lau, The University of Hong Kong, China

 

 ---- "Research Issues in Adapting Computing to Small Devices"

 

Abstract of the Speech

Biographical details of the Speaker

 

The Slides of the Speech

  Prof. Kurt Rothermel, Universitat Stuttgart, Germany
  ---- "Mobile Context-aware Systems - Linking the Physical and Digital World"
 

Abstract of the Speech

Biographical detail of the Speaker

 

The Slides of the Speech

 

 

 
"Balancing System Resource Supply and Demand for Cost Effective High Performance Computing"

Large and modern high-end systems are typically built by commodity processors, which can unfortunately provide unbalanced computing resources with oversupplied CPU cycles and increasingly long latency of both local and remote data accesses. Sustained performance of data-intensive applications in these systems are often low due to low execution efficiency. The imbalance is mainly caused by two dramatic technology changes. First, the speed gaps between the CPU and the memory and the I/O storage continue to grow. Second, the latency improvement significantly lags behind the bandwidth improvement in any pair connection of computer and distributed systems. Recently, we have started to face a new processor technology challenge of changing its single execution core to multicores.

In this talk, I will overview the imbalance problem and its challenges for us to build high end systems. I will also present several research directions to balance system resource supply and demand for cost effective high performance computing. The targeted systems include multicore processors, tightly-couple multiprocessor systems, large loosely-coupled and cluster based multicomputer systems, and globally connected grid systems.

Back to Top

Biographical details of Prof. Xiaodong Zhang

Xiaodong Zhang is becoming the Robert M. Critchfield Professor in Engineering, and the Head of the Computer Science and Engineering Department at the Ohio State University. He is leaving the position of the Lettie Pate Evans Professor of Computer Science and the Department Chair at the College of William and Mary.

His research interests are in the areas of high performance and distributed systems. Several innovations and research results of his Laboratory have been adopted or being developed in commercial products and open source systems, improving the memory controller in Sun UltraSPARC processors, and advancing the memory management system in the Linux kernel. He was the Program Director of Advanced Computational Research at the US National Science Foundation from 2001 to 2004. He has served on the Editorial Board of IEEE Transactions on Parallel and Distributed Systems, and currently on the Editorial Boards of IEEE Transactions on Computers and IEEE Micro.

Xiaodong Zhang receives his Ph.D. in Computer Science from University of Colorado at Boulder, and his B.S. in Electrical Engineering from Beijing Polytechnic University

Back to Top

"Research Issues in Adapting Computing to Small Devices"

Advances in pervasive and mobile technologies are making computing available to us at anytime anywhere. Availability however does not automatically mean it is in a form that implies ease of use. Usability in the mobile world amounts to a set of problems that are not so much precedented in the history of computing. Handheld mobile devices that are thin-lean-mean for instance present challenges that require fundamental changes in the way computation is carried out, its architecture, or its supporting environment. A practical goal is to minimize these changes, which calls for automatic or semi-automatic adaptation of existent computing to the small devices.

We discuss the issues and research challenges of "X adapting to Y", where X includes content, data, code, computation, GUI, and so on, and the changes in semantics and/or syntax due to the adaptation are to satisfy the constraints of Y. Some experiments we have carred out for content and code adaptation provide some useful illustration.

Back to Top

Biographical details of Prof. Francis Lau

Francis C.M. Lau received his PhD in Computer Science from University of Waterloo in Canada in 1986. He has been a faculty member of the Department of Computer Science, The University of Hong Kong since 1987, of which he is now the head of department.

His research specializes in operating systems, parallel/distributed systems, and pervasive/mobile computing. He created the Systems Research Group (SRG) which has graduated many MPhils and PhDs over the years. The group has produced several Java-based software packages which have earned a reputation for the group. SRG is also responsible for the Hong Kong University Grid Point which is a key node of the China National Grid as well as an important gateway to the Asia-Pacific Grid. One of the focuses of his current research is to tackle some of the fundamental problems associated with the realization of the "pervasive grid".

Back to Top

"Mobile Context-aware Systems - Linking the Physical and Digital World"

The rapid miniaturisation and decline in prices of computer, communication and sensor technology give rise to a number of interesting developments, such as multifunctional mobile devices, sensor platforms embedded into everyday things, and sensor nodes organised in a wireless network. Those systems can capture and process sensory data and communicate this information to other peers in their proximity or to an existing server infrastructure. The sensory data are fed into spatio-temporal models of the physical world, which build the basis for the promising class of context-aware applications.

Based on these developments it can be anticipated that there will be billions of sensor systems in our physical environment near future. Consequently, we envision most of the future applications to be context-aware, sharing highly dynamic digital world models offered by a large number of content providers. Obviously, the realisation of this vision will cause a number of both technological and social challenges. Some of these challenges are subject to the research of Nexus, a Centre of Excellence established at University of Stuttgart in the year 2003. In this talk, we present the vision and objectives of Nexus. Moreover, we will discuss some aspects of scaleable context management.
 

Back to Top

Biographical details of Prof. Kurt Rothermel

Kurt Rothermel received his doctoral degree in Computer Science from University of Stuttgart in 1985. From 1986 to 1987 he was ^Post-Doctoral Fellow ̄ at IBM Almaden Research Center in San Jose, U.S.A. and then joined IBM¨s European Networking Center in Heidelberg. Since 1990 he is a Professor for Computer Science at University of Stuttgart. He is Director of Institute of Parallel and Distributed Systems and head of Centre of Excellence Nexus, which is funded by German Science Foundation (DFG) and is conducting research in the area of mobile context-aware systems. His research interest is in the field on distributed and mobile systems, where he contributed more than 120 publications.

Back to Top