Are Magnus Bruaset

Modern software tools for scientific computing

The purpose of this text is to survey recent advances in the development of software tools for scientific computing. It emphasizes the design of large software codes, computational efficiency, object-oriented programming, reliability of numerical software and parallel computing.

A Comprehensive Set of Tools for Solving Partial Differential Equations; Diffpack

This chapter presents an overview of the functionality in Diffpack, which is a software environment for the numerical solution of partial differential equations. Examples on how object-oriented programming techniques are applied for software design and implementation are provided. In addition, we present a collection of sample Diffpack applications.

Object-oriented design of preconditioned iterative methods in diffpack

As modern programming methodologies migrate from computer science to scientific computing, developers of numerical software are faced with new possibilities and challenges. Based on experiences from an ongoing project that develops C11 software for the solution of partial differential equations, this article has its focus on object-oriented design of iterative solvers for linear systems of equations. Special attention is paid to possible conflicts that have to be resolved in order to achieve a very flexible, yet efficient, code.

A Finite-Element Approach for Calculating Electrical Parameters of Umbilical Cables

This paper presents a systematic approach for calculating electrical per-unit-length parameters of signal cables by the finite-element method. The associated software (UFIELD) has been specifically adapted to umbilical cables used in offshore control applications. Efficient and robust case specification is achieved through a library of predefined, parametrized models for the umbilical elements (signal cables, pipes, armors). Additional features include automated specification of boundary conditions and excitations, adaptive mesh generation, and the handling of twisted elements and layers. The approach is validated against analytical solutions. Application to the Snohvit control umbilical shows excellent agreement with measured signal parameters, for the triad element alone, and for the triad element inside the umbilical. Similar results are obtained for the Ormen Lange quad element.

Object-Oriented Numerics

This chapter is concerned with the use of object-oriented programming techniques for numerical applications, especially in terms of the computer language C++. Through a series of examples we expose some of the strengths and possibilities of object-oriented numerics.

On the stability of relaxed incomplete LU factorizations

When solving large linear systems of equations arising from the discretization of elliptic boundary value problems, a combination of iterative methods and preconditioners based on incomplete LU factorizations is frequently used. Given a model problem with variable coefficients, we investigate a class of incomplete LU factorizations depending on a relaxation parameter. We show that the associated preconditioner and the factorization itself both are numerically stable. The theoretical results are complemented by numerical experiments.

On the Numerical Efficiency of C++ in Scientific Computing

We investigate the relative efficiency of C++ and C code versus FORTRAN 77 code through numerical experiments conducted on a range of computer platforms. The problem areas cover basic linear algebra and finite element solution of porous media fluid flow and species transport problems. The C++ codes are short and make extensive use of Diffpack, a generic library based on object-oriented programming techniques, while the FORTRAN and C programs are either based on vendor supplied numerical libraries or written and tuned particularly for the test problem. Challenges encountered in optimizing C++ codes and the efficiency of dynamic memory handling in C++ are also addressed.

Increasing the efficiency and reliability of software development for systems of PDEs

In this chapter we address an object-oriented strategy for the development of software solving systems of partial differential equations. The proposed development strategy encourages heavy reuse of modules capable of solving the involved subproblems. Using class inheritance for successive refinement of the involved solvers, the complexity of the overall model is increased stepwise, layer by layer. In addition to the obvious advantage of code reuse and modular testing, this approach allows the developer to pull the pieces apart at any time for individual verification.

Numerical Solution of PDEs on Parallel Computers Utilizing Sequential Simulators

We propose a strategy, based on domain decomposition methods, for parallelizing existing sequential simulators for solving partial differential equations. Using an object-oriented programming framework, high-level parallelizations can be done in an efficient and systematical way. Concrete case studies, including numerical experiments, are provided to further illustrate this parallelization strategy.

Simula Research Laboratory

The Simula Research Laboratory, located just outside Oslo in Norway, is rightly famed as a highly successful research facility, despite being, at only eight years old, a very young institution. This fascinating book tells the history of Simula, detailing the culture and values that have been the guiding principles of the laboratory throughout its existence. Dedicated to tackling scientific challenges of genuine social importance, the laboratory undertakes important research with long-term implications in networks, computing and software engineering, which includes offering a Centre of Excellence in biomedical computing. The text covers every angle: the laboratorys background and political context, the work it has completed, its approach to the education and development of its researchers, and the organisations efforts to promote the application of its research results. The key research projects and results are covered in separate chapters, each of which is preceded by a thumbnail description of the project and how the research is useful for society. As well as explanatory text, the other chapters feature interviews with major players in the facilitys history, both inside Simula, among its collaborators, and in Norwegian society at large.