Keynote and Invited Speakers
|
|||
Keynote and Invited Speakers |
|
Talk 1: "e-Science, Cyberinfrastructure and Web Service Grids"
EPSRC and University of Southampton Tony Hey is Professor of Computation at the University of Southampton and has been Head of the Department of Electronics and Computer Science and Dean of Engineering and Applied Science at Southampton. From March 31st 2001, he has been seconded to the EPSRC and DTI as Director of the UK's Core e-Science Programme. He is a Fellow of the Royal Academy of Engineering (RAE), the British Computer Society (BCS), and the Institution of Electrical Engineers (IEE) and Member of the Institute of Electrical and Electronics Engineers (IEEE). Professor Hey is European editor of the journal 'Concurrency and Computation: Practice and Experience' and is on the organising committee of many international conferences. Professor Hey has worked in the field of parallel and distributed computing since the early 1980's. He was instrumental in the development of the MPI message-passing standard and in the Genesis Distributed Memory Parallel Benchmark suite. In 1991, he founded the Southampton Parallel Applications Centre in 1991 that has played a leading technology transfer role in Europe and the UK in collaborative industrial projects. His personal research interests are concerned with performance engineering for Grid applications but he also retains an interest in experimental explorations of quantum computing and quantum information theory. As the Director of the UK e-Science Programme, Tony Hey is currently excited by the vision of the increasingly global scientific collaborations being enabled by the development of the next generation 'Grid' middleware. The successful development of the Grid will have profound implications for industry and he is much involved with industry in the move towards OpenSource/OpenStandard Grid software. Tony Hey is also the author of two popular science books: 'The Quantum Universe' and 'Einstein's Mirror'. Most recently he edited the 'Feynman Lectures on Computation' for publication, and a companion volume entitled 'Feynman and Computation'.
http://www.rcuk.ac.uk/escience/
Talk 2: "Experiences with System X"
Director, Terascale Computing Facility System X was conceived in March 2003, designed in July 2003 and by October it had achieved a sustained performance of 10.28 Teraflops, making it the third fastest supercomputer in the world at the time and the seventh fastest today. System X has several novel features. First, it is based on an Apple G5 platform with IBM PowerPC 970 64-bit CPUs. Secondly, it uses a high performance switched communications fabric called Infiniband. Finally, System X is cooled by a hybrid liquid-air cooling system. In this talk, I will present the motivation for System X, its architecture, operational experiences and the challenges faced in building, deploying and maintaining a largescale supercomputer. Bio Srinidhi Varadarajan received his Ph.D. in Computer Science from the State University of New York, Stony Brook in 2000. He presently serves as the Director of the Terascale Computing Facility at Virginia Tech and as an Assistant Professor in the Department of Computer Science. Dr. Varadarajan is the recipient of the ComputerWorld Honors Award in the Science Category 2004, a CAREER award from the National Science Foundation, the Egg Factory Technology Innovation award and a Faculty Fellow award from the College of Engineering, Virginia Tech. He was also honored by the MIT Technology Review in 2004 as one of the top 100 (TR100) young innovators transforming technology. Dr. Varadarajan's research is focused on transparent fault tolerance for massively parallel supercomputers, scalable network emulation, compiler directed strategies for flexible data sharing models and routing algorithms for backbone IP networks. In the area of transparent fault tolerance his work concentrates on developing incremental checkpointing, recovery and migration algorithms. His research in network emulation is focused on building a distributed system that can scale to emulate hundreds of thousands of virtual nodes. This work involves research on several areas, including compiler directed mechanisms for transparent generation of reentrant code from non-reentrant sources, automatic checkpointing and recovery, code migration, dynamic load balancing and 3D environments for network traffic visualization. In the area of routing algorithms, he is exploring the use of AI techniques such as reinforced learning for use in a probabilistic framework for multi-path routing protocols. Dr. Varadarajan is the architect of System X, the third fastest supercomputer in the world located at the Terascale Computing Facility at Virginia Tech.
Industry Keynote: "Enabling robust loosely coupled systems using WS-Agreement" This talk is organised in collaboration with Workshop 9 on Agent-based Grid Economics
IBM TJ Watson Research Center, WS-Agreement is a specification defined in the Global Grid Forum specifying a language for service agreements and interfaces to establish and monitor them. Agreements are central instruments for dynamic and loosely coupled distributed systems that are not centrally managed, such as the Grid. Agreements allow an organizational domain to reserve and use service capacity from another organizational domain. Agreements can be made on different levels: Resource agreements define how much of specific resource types such computing nodes or storage can be used by the service customer. Service agreements define the service to be performed in terms of particular objectives to be achieved. For example a job executing a given application with input data of a particular size must be completed by a specific time. Mixes and combinations are possible. Typically resource-level agreements are more easy to implement by a service provider while service-level agreements are often more desirable by service customers. WS-Agreement addresses resource and service-level scenarios. Using service description and guarantee terms, providers and customers can describe the terms of the service and also define the business value of guarantees, e.g., by defining penalties for non-compliance. The monitoring interface enables to check on agreement status during runtime and take corrective action when necessary, e.g., by acquiring resource from a different provider. This Industry Keynote will introduce the main elements of WS-Agreements and illustrate its use in different application contexts such as job scheduling or providing of transactional (Web) services using specific application examples. Based on this introduction we will discuss how WS-Agreement based systems can lead us beyond best effort-based approaches to loosely coupled systems towards more robust environments providing quality-of-service guarantees, and hence become more viable in commercial environments. Bio Heiko Ludwig is a Research Staff Member at IBM's T.J. Watson Research Center, where he was a visiting scientist since June 2001. As a member of the Internet Infrastructure and Computing Utilities department he works in the field of autonomic service-oriented computing, primarily addressing issues of dynamic resource acquisition and quality of service management in cross-domain service-oriented architectures, including the Grid. This includes contributions to policy representation and management, agreement representation and the role of agreements in service binding and resource acquisition. He represents IBM in the GGF GRAAP working group, working on the WS-Agreement specification. Prior to that he was a Research Staff Member at IBM's Zurich Research Laboratory where he was working on cross-organisational distributed systems, service outsourcing, electronic contracts, outsourcing-related security aspects, and service modelling. From 1992 to 1996 he was a research and teaching assistant at the department of Office Automation at the Otto-Friedrich University Bamberg, Germany. During that time he worked on the integration of workflow and collaborative applications. He holds a Master's (Diplom) degree (1992) and a PhD (1997) in computer science and business administration from Otto-Friedrich University Bamberg, Germany. He published a book and multiple book chapters, numerous journal articles and conference papers, acted in program committees, and organized workshops in the area of CSCW, workflow management, e-business infrastructures, contracts and policies. He guest edited multiple journal issues and serves on editorial boards.
Cluster Security Invited Speaker:
"SSH and Cascade Security Failures: This talk is organised in collaboration with Workshop 2 on Cluster Security
Over the past year, attacks on SSH have compromised major supercomputing facilities, educational institutions, and national laboratories. These attacks have proven inadequate our current mechanisms for authenticating users and then isolating them from each other, especially when viewed in the context of shared GRID systems. I will describe the mechanisms that have been used to attack SSH and other remote execution mechanisms, and then present data to help explain why these attacks have been so successful. I will describe countermeasures that can be used to make SSH more resilient to some of these attacks. However, other attacks require us to rethink our entire approach to authenticating ourselves to remote hosts and services and authorizing other hosts to perform tasks on our behalf. Bio Stuart Schechter is a member of Information Assurance Group at MIT Lincoln Laboratory. His technical research encompasses areas including attack modeling, prevention, and detection, with a focus on malicious code. He also studies economic approaches to measuring and improving computer security. Dr. Schechter received his Ph.D. in Computer Science from Harvard University in 2004. Stuart remains part of the Harvard community through his role as a Resident Tutor in Leverett House. He received his Bachelors, also in Computer Science, in 1996 from The Ohio State University College of Engineering. Now that he's at MIT, Stuart refers to the igloos he builds in his free time by number, rather than by name.
|
Professor Tony Hey 
Professor Srinidhi Varadarajan 
Dr Heiko Ludwig 
Dr. Stuart Schechter