Invited Talks

The Role of Virtual Machines in Future Computer Systems
Jim Smith, University of Wisconsin-Madison, USA
Abstract: Virtual machines can enhance computer systems in a number of ways, including improved security, flexibility, fault tolerance, power efficiency, and performance. Virtualization can be done at the system level and the process level. Virtual machines can support high level languages as in Java, or can be implemented using a low level co-designed paradigm as in the Transmeta Crusoe. This talk will survey the spectrum of virtual machines and discuss important design problems and research issues. Special attention will be given to co-designed VMs and their application to performance- and power-efficient microprocessor design.

High Performance Computing using Reconfigurable Hardware
Viktor Prasanna, University of Southern California, USA
Abstract: Reconfigurable hardware such as FPGAs(Field Programmable Gate Arrays) offer the potential of energy efficiency and superior time performance. We first outline an algorithm level design methodology for optimising the energy performance of FPGA based implementations and introduce "malleable algorithms" as an approach to develop energy efficient soft IP(Intellectual Property) cores for FPGAs. We consider several kernels in adaptive signal processing in Software Defined Radio (SDR) and demonstrate energy and time efficient algorithms and implementations for these on FPGAs. We also develop algorithmic optimizations and demonstrate the suitability of FPGAs for floating point intensive computations. The performance of FPGAs is also compared against those of state-of-the-art embedded processors, general purpose processors, and DSPs.

The Microprocessor of the Year 2014: Do Pentium 4, Pentium M, and Power 5 provide any hints?
Yale N. Patt, The University of Texas at Austin, EUA
Abstract: Moore's Law continues to thrive, -- one billion transistors on each silicon die with a 10 GHz clock is close at hand. But Pentium 4 follow-ons, we are told, have been canceled. On the other hand, Intel has made some provocative statements with Pentium M. IBM has done likewise with Power 5. Can we continue to harness the raw technology available on the chip? If so, where will we look for solutions, and what will this mean for the microprocessor for high performance implementations ten years from now?

The Gridbus Toolkit for Building and Deploying eScience Applications on Utility Grids
Rajkumar Buyya, University of Melbourne, Australia
Abstract: Grid Computing is emerging as a next-generation parallel and distributed computing paradigm driven by the Internet, Web services technologies, and service-oriented computing architectures. Grids enable the sharing, selection, and aggregation of geographically distributed resources, such as computers (PCs, workstations, clusters, supercomputers), data sources, and scientific instruments, for solving large-scale problems in science, engineering, and commerce. To realise the full potential of computing Grids, a number of projects, both within Australia and around the world, have been making steady progress in the design, development, and deployment of Grid technologies and applications.

The notion of utility computing, which enables the leasing of information technology (IT) services on demand, is gaining wide attention due to its cost effective computing nature. Hence, the need for technologies that unify the notion of Service-Oriented Architectures, Grid and utility computing; and empower both service providers and consumers to operate based on their individual needs is rapidly growing.

The Gridbus Project is engaged in the design and development of grid middleware technologies to support eScience and eBusiness applications. They include visual Grid application development tools for rapid creation of distributed applications, competitive economy-based Grid scheduler, cooperative economy-based cluster scheduler, Web-services based Grid market directory (GMD), Grid accounting services, Gridscape for creation of dynamic and interactive testbed portals, G-monitor portal for web-based management of Grid applications execution, and the widely used GridSim toolkit for performance evaluation. Recently, the Gridbus Project has developed a .NET-based clustering and Grid web services framework to support the integration of both Windows and Unix-class resources for Grid computing.

This presentation identifies Grid synergies, challenges, and opportunities; and presents architectural framework, methodologies, and Gridbus technologies that are being developed to realise scalable utility annd service-oriented computing. We present/demonstrate (live, if Internet access is available) the usage of Grid tools in composition and distributed execution of data-intensive applications (e.g., molecular docking, brain activity analysis, and high-energy physics) on the World-Wide Grid testbed resources located in Australia, Asia, Europe, North America, and South America. The presentation concludes by highlighting sociological and intellectual implications of this new Internet-based computing paradigm and its impact on the marketplace. For further information on the Gridbus Project, please visit -