U.S.-Finland Workshop on Advances in Scientific Computing with Emphasis on Linear Problems; Helsinki, Finland; August,1989

该奖项为大约12名美国计算机科学家提供了部分支持，以参加在芬兰举行的关于大规模科学计算技术状况的联合研讨会。会议由斯坦福大学的Gene Golub教授和赫尔辛基工业大学数学系的Olavi Nevanlinna教授组织。它将与SIAM（工业与应用数学学会）北欧分会的年度会议同时举行。芬兰在数值分析方面有很强的传统，特别是在混合有限元、自由边界问题数值和积分方程数值等方面。然而，那里的数学家在大规模计算方面的经验相对较少。芬兰国家计算机中心刚刚获得了它的第一台超级计算机，这极大地扩展了芬兰科学计算的潜力。研讨会将集中讨论一般意义上的大规模计算，特别是数值线性代数，并有望促进美国和芬兰在这些领域的合作。美国参与者应该从芬兰小组在各种情况下应用有限元技术的经验中获益良多，包括一些新颖的应用。随着新的计算机体系结构和更强大的计算机的出现，有必要重新考虑数值分析与计算机体系结构之间的联系。许多过去被抛弃的算法现在又回到了前沿。这些算法对计算有很大的影响，可以在设计求解方法的制定中起到一定的作用。本次研讨会将进一步集中讨论问题的提出和用于解决这些问题的方法之间的关系。

This award provides partial support for approximately twelve U.S. computer scientists to participate in a joint workshop in Finland on the state of the art in large scale scientific computing. The meeting was organized by Professor Gene Golub of Stanford University and Professor Olavi Nevanlinna of the Mathematics Department of Helsinki University of Technology. It will be held in conjunction with the annual meeting of the Nordic Section of SIAM (Society for Industrial and Applied Mathematics.) Finland has a strong tradition in numerical analysis, particularly in such topics as mixed finite elements, numerics on free boundary problems, and numerics on integral equations. However, mathematicians there have relatively little experience in large scale computing. The Finnish State Computer Center has just acquired its first supercomputer, which has greatly expanded the potential for scientific computing in Finland. The workshop will focus on large scale computing in general and numerical linear algebra in particular, and is expected to stimulate U.S.-Finnish cooperation in these areas. The U.S. participants should benefit greatly from the experience of the Finnish group in applying finite element techiques in a wide variety of situations, including some novel applications. With the advent of new computer architectures and more powerful computers, it is necessary to consider anew the link between numerical analysis and computer architectures. Many algorithms which have been discarded in the past are now returning to the forefront. These algorithms can greatly affect a computation and can play a role in the formulation of the design of the solution method. This workshop will focus further attention on the relation between the problem formulation and the methods which are used for solving these problems.