Software Optimization for High Performance Computing: Creating Faster Applications

Description

Prentice Hall, 2000 — 377 p.This book is written for programmers who want to get high performance from the software they write. The optimization techniques discussed are applicable to all computers, but are of most interest to designers of software for high performance computers, since they are most concerned with high performance. The main focus is on UNIX, though since Windows NT is now being used on high performance servers, there is limited coverage on this. Single processor and parallel performance is discussed.ContentsIntroduction.Hardware Overview-Your Work Area.Processors: The Core of High Performance Computing.Data Storage.An Overview of Parallel Processing.Software Techniques-The Tools.How the Compiler Can Help and Hinder Performances.Predicting and Measuring Performance.Is High Performance Computing Language DependentParallel Processing-An Algorithmic Approach.Applications-Using The Tools.High Performance Libraries.Mathematical Kernels: The Building Blocks of High Performance.Faster Solutions for Systems of Equations.High Performance Algorithms and Approaches for Signal Processing.Index.