Thanks for visiting Scientific Computing World.

You're trying to access an editorial feature that is only available to logged in, registered users of Scientific Computing World. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

ISC 08 unlocks keynotes

Share this on social media:

The International Supercomputing Conference (ISC’08) has released details of the keynotes and pre-conference talks to be held at the event next month.

Two of the keynote talks will centre on different ways to bring High-Performance Computing (HPC) to the masses, with the other focusing on the effects of HPC on molecular dynamics simulations.

Conference chair Professor Hans Werner Meuer of the University of Mannheim, said: ‘These [keynote] speakers are people who are changing the landscape in computing and who can provide insight on everything from the evolution of grid infrastructure to the transformation of computing through multicore platforms to scientific discoveries enabled by new infrastructure and algorithms.’ 

First up is Professor Satoshi Matsuoka of the Tokyo Institute of Technology Global Scientific Information and Computing Center, who has been involved in grid computing research for more than a decade.

He is one of the leaders of the Japanese National Research Grid Initiative, a project that aims to create middleware for next-generation science cyberinfrastructure, and was the technical lead in the construction of TSUBAME, the fastest supercomputer in Asia with a peak performance of 85 teraflops. Matsuoka will open the conference technical program on 18 June with a talk that will highlight the changing face of grid computing. In the era of ubiquitous parallel processors and a proliferation of Web 2.0 technologies, grid infrastructure could become more centralised and supercomputers could become much more like internet-based resources rather than elite resources available only to a select audience, according to Matsuoka. His talk will discuss future multi-petascale grids, which could become massive resource nodes supporting service-oriented architectures rather than vast distributions of smaller resources. This evolution of supercomputing grid could finally make supercomputing available to the masses. 

Justin Rattner, Intel Corporation’s chief technology officer, will present another view of supercomputing for the masses within the second keynote on multicore/many core platforms on 19 June. Rattner will address the transformation of the computing world made possible by the spread of multicore processors and how multicore revolution could usher in a new internet age characterised by realistic graphics and personalised information spaces. He will address Intel’s terascale research programme and the challenge of scaling multicore architectures to integrate programmable cores and fixed-function accelerators, flexible cache and memory hierarchy, and high-bandwidth on-die networks to ensure high throughput. 

Finally, Professor John Salmon of D. E. Shaw Research, New York, USA, will present the keynote on 20 June, which will focus on the effects of high performance computing on molecular dynamics (MD) simulations.

Dr Salmon is involved with algorithm design, software development, and computer architecture design. He will discuss Anton, a specialised, massively parallel computing system being built at Shaw which, when completed later this year, should be capable of executing millisecond-scale classical molecular dynamics (MD) simulations of one or more proteins at the atomic level of detail.

MD simulations at this timescale are about three orders of magnitude beyond the duration of the longest current MD simulations. The ability to simulate this level of molecular activity could provide a new tool for drug discovery. The conference will be held 17 - 20 June at the International Congress Center in Dresden. The conference, which is conducted in English, includes three days of technical sessions, pre-conference sessions aimed at the automotive industry and scientific researchers, poster and birds-of-a-feather sessions, and after-hours networking opportunities.

ISC’08  will also feature three one-day preconference programs: HPC Requirements for the Automotive Industry; Scientific Sessions I and II; and Efficient Operation of Clusters in Computing Centres.

The pre-conference sessions will take place the afternoon of 17 June, the day before the ISC technical program begins. 

The automotive session will focus on HPC trends in the automotive industry and on the crucial role played by HPC systems in helping automakers succeed in the increasingly competitive global market. Discussion topics will include the challenges faced by managers of automotive industry IT environments and computing infrastructures best suited to meet the needs of the automotive industry. High performance computing applications and visual simulations used to improve automotive design and reduce costs also will be featured.

The new preconference session on clusters in computing centres will provide IT managers, computing system manufacturers and computing center users with in-depth information about the important issues related to developing and maintaining a computing center of commodity-based clusters. Speakers will discuss performance and optimisation issues, software challenges, developments in security, data management and energy consumption, and more.

The session is designed to help businesses, universities and research centres choose clusters that best fit their needs and optimize their existing clusters.

Scientific Day sessions will showcase the best scientific papers submitted through the conference’s call for papers. Two papers will be chosen by the ISC Scientific Program Committee to receive the ISC Award. Topics to be covered will be:

 

  • Advances in very large-scale applications for science and engineering modeling, simulation and design (or capability computing)
  • Successful implementation and deployment of large-scale, grid-enabled systems
  • Computation pipelines and workflows for the life sciences, especially computational biology and biological data integration
  • High performance computer architectures
  • Data-intensive and I/O computing
  • Scalable tools for performance analysis and tuning for hundreds and thousands of PEs
  • High-speed system area networks for large computer system architectures
  • Measurement and modeling of grid middleware and applications