Guide to Scientific Computing in C++

1. Front Matter

   Pages i-xiv

   PDF

2. Getting Started

   • Joe Pitt-Francis, Jonathan Whiteley

   Pages 1-26

3. Flow of Control

   • Joe Pitt-Francis, Jonathan Whiteley

   Pages 27-45

4. File Input and Output

   • Joe Pitt-Francis, Jonathan Whiteley

   Pages 47-58

5. Pointers

   • Joe Pitt-Francis, Jonathan Whiteley

   Pages 59-73

6. Blocks, Functions and Reference Variables

   • Joe Pitt-Francis, Jonathan Whiteley

   Pages 75-97

7. An Introduction to Classes

   • Joe Pitt-Francis, Jonathan Whiteley

   Pages 99-128

8. Inheritance and Derived Classes

   • Joe Pitt-Francis, Jonathan Whiteley

   Pages 129-143

9. Templates

   • Joe Pitt-Francis, Jonathan Whiteley

   Pages 145-161

10. Errors, Exceptions and Testing

    • Joe Pitt-Francis, Jonathan Whiteley

    Pages 163-181

11. Developing Classes for Linear Algebra Calculations

    • Joe Pitt-Francis, Jonathan Whiteley

    Pages 183-196

12. An Introduction to Parallel Programming Using MPI

    • Joe Pitt-Francis, Jonathan Whiteley

    Pages 197-224

13. Designing Object-Oriented Numerical Libraries

    • Joe Pitt-Francis, Jonathan Whiteley

    Pages 225-244

14. Back Matter

    Pages 245-287

    PDF

This simple-to-follow textbook/reference provides an invaluable guide to object-oriented C++ programming for scientific computing. Through a series of clear and concise discussions, the key features most useful to the novice programmer are explored, enabling the reader to quickly master the basics and build the confidence to investigate less well-used features when needed. The text presents a hands-on approach that emphasizes the benefits of learning by example, stressing the importance of a clear programming style to minimise the introduction of errors into the code, and offering an extensive selection of practice exercises.

This updated and enhanced new edition includes additional material on software testing, and on some new features introduced in modern C++ standards such as C++11.

Topics and features: presents a practical treatment of the C++ programming language for applications in scientific computing; reviews the essentials of procedural programming in C++, covering variables, flow of control, input and output, pointers, functions and reference variables; introduces the concept of classes, showcasing the main features of object-orientation, and discusses such advanced C++ features as templates and exceptions; examines the development of a collection of classes for linear algebra calculations, and presents an introduction to parallel computing using MPI; describes how to construct an object-oriented library for solving second order differential equations; contains appendices reviewing linear algebra and useful programming constructs, together with solutions to selected exercises; provides exercises and programming tips at the end of every chapter, and supporting code at an associated website.

This accessible textbook is a “must-read” for programmers of all levels of expertise. Basic familiarity with concepts such as operations between vectors and matrices, and the Newton-Raphson method for finding the roots of non-linear equations, would be an advantage, but extensive knowledge of the underlying mathematics is not assumed.

• C++ Programming
• MPI
• Numerical Methods
• Object-Oriented Programming
• Parallel Programming
• Scientific Computing

• University of Oxford, Oxford, United Kingdom

  Joe Pitt-Francis, Jonathan Whiteley

Dr. Joe Pitt-Francis is a Departmental Lecturer at the Department of Computer Science and teaches Computer Science at Exeter College, the University of Oxford, UK.

Prof. Jonathan Whiteley is Associate Professor at the Department of Computer Science and Governing Body Fellow of Linacre College, the University of Oxford, UK.