• Dr. William Pulleyblank, IBM Research, Proteins, Petaflops and Algorithms.

    Computational biology is  an important, rapidly growing  area of deep computing.  The protein folding problem is one of the most intriguing problems - how does a protein  form a three dimensional structure when it is placed in water?   Modeling  this process goes far beyond the capabilities of current supercomputers.  I will discuss the problem as well as different solution approaches currently being tried.  I will also discuss a project called Blue Gene which will build a petaflop scale supercomputer suitable for one approach to this problem within the next three years.

     

  • Dr. Ian Foster, ANL, An Open Grid Services Architecture for Distributed Systems Integration.

    In both e-business and e-science, we often need to integrate services across distributed, heterogeneous, dynamic "virtual organizations" formed from the disparate resources within a single enterprise and/or via external resource sharing and service provider relationships. This integration can be technically challenging due to the need to achieve various qualities of service when running on top of different native platforms. We present an Open Grid Services Architecture that addresses these challenges.  Building on concepts and technologies from the Grid and Web services communities, this architecture defines a uniform exposed service semantics (the Grid service); defines standard mechanisms for creating and discovering transient Grid service instances; provides location transparency and multiple binding protocols for service instances; and supports mapping services for integration with underlying native platform facilities. The Open Grid Service Architecture defines, in terms of Web Services Description Language interfaces, mechanisms required for creating and composing sophisticated distributed systems, including lifetime management, reliable remote invocation, change management, credential management, and notification. Our presentation describes how Globus Toolkit mechanisms can be used to implement a service oriented architecture, explains how Grid functionality can be incorporated into a Web services framework, and illustrates how our architecture can be applied as a basis for distributed system integration--within and across organizational domains.

     

  • Dr. John Gustafson, Sun Microsystems, The Cost of the Single System Image.

    We now have reached the point where parallel systems scale linearly in price if they are simply a collection of processors, but the price grows as the square of the number of processors if we require that the ensemble presents a single system image. Is it worth it? Oddly enough, the free market seems to think so. This suggests that we might have to reconsider our traditional models of parallel efficiency and speedup, and find better ways to quantify when the single system image is truly necessary.