Content

1. Introduction
- Characteristics of scientific software

2. Working with real numbers
- deeper study of IEEE 754 (not limited to number representation, but also dealing with less known aspects such as calculation rules and floating point exceptions)
- multiple-precision arithmetic
- interval arithmetic

3. Languages for scientific applications
- Comparative study of higher programming languages from the point of view of scientific calculations (support of IEEE754, floating point exceptions, matrix representation, intrinsic speed limitations...)
- Object-oriented design in non-pure OO-languages
- Efficiency as attention point during the design and implementation of software
- More thorough study of languages designed for calculations.

4. Working with real computers
- Performance improvement on 1 processor (with multiple calculating units)
- Benchmarking

5. Working with real algorithms
- Testing and evaluating numerical software
- The compromise between reliability and efficiency
- Implementing machine-dependence in portable software

6. Directions in mathematical software
- Overview of various sources
- Use of libraries versus 'tools'

Course material

Study cost: 51-75 euros (The information about the study costs as stated here gives an indication and only represents the costs for purchasing new materials. There might be some electronic or second-hand copies available as well. You can use LIMO to check whether the textbook is available in the library. Any potential printing costs and optional course material are not included in this price.)

Textbook/articles and literature/course text/Toledo
- Book "Accuracy and reliability in scientific software", IFIP WG 2.5 Project 68.
- Educating programmes by Cornell Theory Center

Content

During the guided exercise sessions, the students become familiar with different aspects of scientific software. The guided exercise sessions serve as a springboard for the home work assignments.

Course material

Manuals Fortran 95/2003 and C++.