
2019 / xii + 251 pages / Softcover / 9781611975550 / List Price $44.00 / SIAM Member Price $30.80 / Order Code: SE29
Keywords: Software Errors, Software Failures, Computational Science, Scientific Computing, Numerical Methods
Contents
List of Excursions;
Preface;
Chapter 1: Introduction;
Chapter 2: Machine Numbers, Precision, and Rounding Errors;
Chapter 3: Mathematical Modeling and Discretization;
Chapter 4: Design of Control Systems;
Chapter 5: Synchronization and Scheduling;
Chapter 6: SoftwareHardware Interplay;
Chapter 7: Complexity;
Chapter 8: Appendix;
Index.
In scientific computing (also known as computational science), advanced computing capabilities are used to solve complex problems. This selfcontained book describes and analyzes reported software failures related to the major topics within scientific computing: mathematical modeling of phenomena; numerical analysis (number representation, rounding, conditioning); mathematical aspects and complexity of algorithms, systems, or software; concurrent computing (parallelization, scheduling, synchronization); and numerical data (such as input of data and design of control logic).
Readers will find lists of related, interesting bugs, MATLAB examples, and "excursions" that provide necessary background, as well as an indepth analysis of various aspects of the selected bugs. Illustrative examples of numerical principles such as machine numbers, rounding errors, condition numbers, and complexity are also included.
This book is intended for students, teachers, and researchers in scientific computing, computer science, and applied mathematics. It is also an entertaining and motivating introduction for those with a minimum background in mathematics or computer science. Bits and Bugs can be used for courses in numerical analysis, numerical methods in linear algebra/ODEs/PDEs, introductory software engineering, introductory scientific computing, and parallel programming.
About the Authors
Thomas Huckle has been professor of scientific computing at the Technical University of Munich since 1995. He has conducted research at Stanford University (1993), UCLA (2000), and Leeds University (2006) and in Como, Italy (2011). His research focuses on topics in numerical linear algebra, such as preconditioning, regularization methods, multigrid methods, structured matrices, and tensors, and algorithms for parallel and highperformance computing. Since 1999, he has maintained a web page called Collection of Software Bugs.
Tobias Neckel is senior researcher in scientific computing at the Technical University of Munich and has conducted research at the École Polytechnique, France (2003), the Tokyo Institute of Technology (2008), and the Australian National University (2017). His research interests include the numerical solution of partial differential equations, hierarchic and adaptive methods, uncertainty quantification, and various aspects of highperformance computing.
ISBN 9781611975550