High performance computing

the long version

This book shows by example how to solve complex scientific problems with programs that run on high performance computers. Combining case studies from a variety of problem domains and written by experts in those domains, it shows how to map or transform an abstract problem into a concrete solution that executes rapidly and efficiently on available high performance hardware. Each chapter describes some technical or engineering problem to which computers are commonly applied, then leads readers through the choice and development of appropriate algorithms and through an evaluation of resulting implementations. Although each chapter centers on a single application area, they all focus on general techniques useful for any area rather than on application-specific details. Readers should have some programming experience in a language like FORTRAN or C and some familiarity with computing issues involved in scientific problems; but no expertise in parallel computing is required.