Performance Evaluation of Supercomputers: (Special Topics in Supercomputing)

Although supercomputer systems are faster, and have larger memory hierarchies than other computer systems, such characteristics merely imply the existence of great potential power. How much of that power can be harnessed productively is the central theme of performance evaluation. A set of methods for evaluating the performance of applications on supercomputers has not yet been rigorously defined. This volume is a compilation of research approaches and techniques that are a promising means to that end. The contributions are grouped into three sections. Contributors to Performance looks at applications, algorithms, compilers, operating systems, and memory issues. Measurements and Metrics addresses some of the current techniques and methods of performance evaluation. Among the topics are: the performance monitoring capabilities of the CEDAR system, the methodology behind the Livermore loops, the empirical analysis of system performance, and a software simulator developed in connection with the RP3 project. Methods, Models, and Directions looks at ways of establishing a general and theoretical framework for supercomputer performance evaluation.

Table of Contents:
Contributors to Performance. Supercomputer Performance Evaluation: The Comparative Analysis of High-Speed Architectures Against Their Applications (J.L. Martin). Using Algorithm Characteristics to Evaluate Parallel Architectures (L.H. Jamieson). Issues in Compiler Performance (J.R.B. Davies). The Effects of Operating Systems on Supercomputer Performance (E. Williams). Measurement and Analysis of Memory Conflicts on Vector Multiprocessors (D.A. Calahan, D.H. Bailey). Measurements and Metrics. Performance Analysis on the CEDAR System (K.A. Gallivan et al.). The Livermore FORTRAN Kernels Test of the Numerical Performance Range (F.H. McMahon). PSIMUL - A System for Parallel Simulation of Parallel Systems (K. So et al.). Problem Related Performance Parameters for Supercomputers (R.W. Hockney). Methods, Models, and Directions. Graph Structured Performance Models (J.C. Browne, A.K. Adiga). Workload Characterization for Supercomputers (M. Calzarossa, G. Serazzi). Empirical Computer Performance Measurement (E.N. Miya). Performance Estimates for Applications: an Algebraic Framework (D. Muller-Wichards). Research Questions for Performance Analysis of Supercomputers (P.J. Denning, G.B. Adams III).