Towards a compilation paradigm for computational applications on the information power grid

The Information Power Grid (IPG) refers to the world-wide infrastructure of computers and their interconnections. We are only at the very beginning of exploring applications and system software that can exploit these resource s effectively and, at the same time, provide adequate interfaces to the pro grammer and end-user. In this paper, we discuss compiler technology that se rves this purpose. The core consists of methods and services that enable an application to adapt flexibly to the dynamically changing nature of the IP G. We present several applications that demonstrate performance opportuniti es in such scenarios. We:then present and evaluate new compiler techniques and services that allow an application to be dynamically re-optimized as th e execution environments change. One very important property of such system support is that the overheads of dynamic re-optimizations are kept small. We will show that this is the case for the presented techniques. The result ing applications can run close to the best performance that could be obtain ed with prior knowledge of the runtime situations. (C) 2000 IMACS. Publishe d by Elsevier Science B.V. All rights reserved.