----- Norsk Informatikkonferanse 1994, Molde 15 November 1994. Side 1 ----- Why Computational Science and Engineering Should be of Interest to Computer Scientists Lasse Natvig Group for Computer Architecture and Design, Department of Computer Systems and Telematics, Norwegian Institute of Technology (NTH), University of Trondheim, N-7034 Trondheim, Norway. (E-mail: Lasse.Natvig@idt.unit.no) Abstract: This paper briefly introduces the field computational science and engineering (CS&E), and is an attempt to get other computer scientists more interested in CS&E related activities. It starts by giving a short outline of the increased international activity in the field. Several of the definitions of CS&E that have been given are presented, with an emphasis on how the field is related to computer science. The role of supercomputers is discussed, and we try to identify important challenges for future CS&E education and research, again with a hope to attract computer scientists. Keywords: computational science and engineering, scientific computing, supercomputing, high performance computing (HPC), education, computer science. 1. Introduction Motivation As part of a committee work on a possible building up on computational science and engineering at the University of Trondheim we have tried to identify what Computational Science and Engineering (CS&E) really is, and how it is related to other more established fields such as mathematics and computer science. During this work, being a computer scientist in a computer science department, the author has to his own surprise found a very low extent of interest in CS&E among computer scientists. This has motivated for writing the paper. It tries to explain computer scientists (including myself) what computational science is, and how it contains a lot of interesting challenges for computer science research. I also argue, perhaps a bit provoking for non-computer scientists, that the computer science community has a responsibility for participating on the CS&E arena to ensure quality. When reading related publications it is clear that the lack of interest in CS&E among computer scientists has been observed by many authors, both nationally and internationally [BW93, Rice93, misc94]. At the University in Trondheim, researchers from the various engineering departments have expressed opinions like; "the computer scientists are playing in their own field ignoring the use of computers to solve important real world problems in science and engineering". This is far from entirely true, but it seems clear that computer science at many places has "grown away" from problem solving in natural sciences and engineering. We try to sketch some possible reasons for this kind of dissatisfaction in Section 5. The paper may be regarded as a small attempt to improve the collaboration between our computer science department and other departments. Increased international activity We have lately seen a world-wide increased activity in high-performance computing in natural sciences and engineering. Initiatives such as HPCC (High Performance Computing and Communication) in the U.S. and HPCN (High Performance Computing and Networking) [HPCN94] in Europe are to a large extent focusing on computational science. A substantial part of the fourth framework programme of the European Union [EC94] is devoted to HPCN.