Wolfgang Bangerth is working on the numerical solution of partial differential equations by the finite element method, in particular for biomedical imaging and other inverse problems. To support these computations, he and two colleagues started the deal.IIproject in 1998, that provides an object-oriented software library implementing the basic elements of finite element programs: handling of meshes in 1d, 2d, and 3d; a wide variety of finite element classes; sequential, parallel, and distributed linear algebra; and input and output in a number of different formats. The foundation of the library is the paradigm that all computations are performed on iterator objects looping over all cells, the faces of cells, or other geometric entities. Through the judicious use of template specialization techniques, codes are written in a dimension-independent way, and are run for a particular space dimension (1d, 2d, or 3d) by mere recompilation. This allows to debug code with relatively short 2d computations, and immediate reuse of the same code for much more demanding 3d applications.

Wolfgang's brainchild, deal.II, has become one of the best known Open Source finite element libraries, with more than 100 downloads and several thousand hits per month on its web page. For its creation, the authors have received the 2007 J. H. Wilkinson Prize for Numerical Software; the library is also part of the SPEC CPU2006 test suite.

Kei Davis, Ph.D. Computing Science (Glasgow), M.Sc. Computation (Oxford), is a is a research scientist in the Performance and Architecture Laboratory at Los Alamos National Laboratory, U.S.A. Current research activities include performance analysis, modeling, prediction, and optimization of physics simulation codes on largest-scale parallel and heterogeneous architectures. Previously he has worked in applying object-oriented and functional language technology to large system design and implementation, scripting, signal processing, parallel discrete-event simulation, and high performance scientific computing. Formerly as research scientist at the Computing Research Laboratory, New Mexico State University, the same technologies were applied to system design and implementation and natural language processing algorithms.

Peter Gottschling works for the Institute for Scientific Computing Research at Dresden University of Technology where he designs and implements next generation scientific C++ software. He is Director for Software Research at Stillwater Supercomputing, Inc. leading the development of the Stillwater Knowledge Engine. He is the author of the Matrix Template Library version 4 (MTL4) that is emerging as the most advanced library in scientific computing combining intuitive mathematical notation with performance in the range of hand-tuned assembler libraries and the latest algorithmic discoveries. By integrating MTL4 into the finite element software AMDiS (Adaptive Multi-Dimensional Simulations), many parts of the latter can be accelerated significantly. MTL4 was recently integrated into FEniCS – one of the largest and most advanced projects in scientific computing. First benchmarks of the FEniCS team demonstrated that sparse matrices can be assembled in MTL4 with half the memory and less than a third of the time compared to the established library uBLAS. Peter Gottschling is co-author of the Parallel Boost Graph Library – the only scalable general-purpose graph library world-wide. At University of Hertfordshire he developed a graph-based library for transformations in automatic differentiation (AD). This library – called Angel – is the most complete implementation of this kind of transformation. Later he integrated Angel during a research visit at the Argonne National Laboratory into the largest AD environment OpenAD where it became a core component. Dr. Gottschling received his Ph.D. Magma Cum Laude in Computer Science from Dresden University of Technology. He is member of ACM and DIN (the German institute for standardization). Peter Gottschling is head of the German delegation in the international ISO committee for the standardization of C++.

René Heinzl studied electrical engineering at the Technische Universität Wien, where he received the degree of Diplomingenieur in 2003 and his PhD in technical sciences in 2007. His research interests include programming paradigms, high performance programming techniques, data structural aspects of scientific computing, performance analysis, process simulation, solid modeling, scientific visualization, algebraic topology, and mesh generation/adaptation.

Bernd Mohr started to design and develop tools for performance analysis of parallel programs already 1987 with his diploma thesis at the University of Erlangen in Germany, and continued this in his Ph.D. work. During a three year PostDoc position at the University of Oregon, he was responsible for the design and implementation of the original TAU performance analysis framework for the parallel programming language pC++. Since 1996 he is a senior scientist at the Juelich Supercomputing Centre of Forschungszentrum Juelich in Germany. Besides being responsible for user support and training in regard to performance tools, he is leading the KOJAK research group on automatic performance analysis of parallel programs which is a joint project with the Innovative Computing Laboratory at the University of Tennessee. He was a founding member and work package leader of the European Community IST working group on automatic performance analysis: APART. He is the author of several dozen conference and journal articles about performance analysis and tuning of parallel programs.

Jörg Nolte is professor for computer science at the Brandenburg University of Technology in Cottbus (Germany) where he holds the chair for distributed systems and operating systems. Prior to that position he was a senior researcher at the Fraunhofer Gesellschaft, Institute for Computer Architecture and Software Technology (FIRST), Berlin. He received his M.S. (Dipl.Inform.) in computer science in 1988 and his Ph.D. (Dr.-Ing.) in 1994, both from the Technical University of Berlin. He was a principal member and finally the deputy head of the PEACE group that developed the operating system for Germany's first massively parallel supercomputer. In 1996 he became the head of FIRST's research group for parallel runtime systems. He was also a member of the EC Europa working group on parallel C++ as well as a steering committee member of ISCOPE, the International Symposium on Computing in Object-Oriented Parallel Environments. Dr. Nolte was also FIRST's representative in the Tsukuba Research Center (TRC) of the Real World Computing Partnership (RWCP) in Tsukuba Science City, Japan. His major research interests are operating systems, middleware and programming languages for parallel, distributed and embedded systems.

Laurent Plagne received his PhD in physics from the University of Grenoble, France. He is a research engineer at Electicité de France, one of largest electricity utilities in Europe. His research interests include parallel PDE solvers, High Performance Computing through Object Oriented and Generic programming. The management of the ever growing complexity of simulation tools in the industrial context is a recurring topic in his work. He co-authored a versatile library for performance assessment (BTL++) and is the main developer of a block-recursive linear algebra library in C++ (LEGOLAS++).

Jörg Striegnitz received his Diploma and Ph.D. in Computer Science from University of Technology at Aachen, Germany. He is now a professor in Theoretical Computer Science and Programming Languages at the University Of Applied Sciences in Regensburg, Germany. His research work includes the integration of programming languages by means of partial evaluation, the application of multiparadigm programming to real world problems, the optimization of programs, and parallel/high performance scientific computing. He authored the FACT! and the EML C++ libraries, that enable programming in a functional style with C++.