On the Efficiency of Superscalar and Vector Computer for Some Problems in Scientific Computing

. Some details of arithmetic of two representatives of computers (a superscalar workstation and a vector uniprocessor) available in the Czech Republic for scientific computing are described. Consequently, their efficiency and precision on a set of linear algebraic tasks solved by different solvers is compared. 1 Introduction One very important question in scientific computing is how to guarantee efficient performance of a given computer on a set of fundamental operations. On the other hand, efficiency on a set of basic, often linear algebra operations, plays usually the key role in the choice of computer for scientific computations. Although a lot of effort has been devoted to developing benchmarks for scientific computing [5], [9], [2] there is still a lack of comparison for computers avilable using a broader spectrum of the important basic algorithms. We devote our attention to some properties of computer arithmetic of two computers which present different philosophy of performing a..