Parallel Algorithmic Techniques for Combinatorial Computation

Parallel computation offers the promise of great improvements in the solution of problems that, if we were restricted to sequential computation, would take so much time that solution would be impractical. There is a drawback to the use of parallel computers, however, and that is that they seem to be harder to program. For this reason, parallel algorithms in practice are often restricted to simple problems such as matrix multiplication. Certainly this is useful, and in fact we shall see later some non-obvious uses of matrix manipulation, but many of the large problems requiring solution are of a more complex nature. In particular, an instance of a problem may be structured as an arbitrary graph or tree, rather than in the regular order of a matrix. In this paper we describe a number of algorithmic techniques that have been developed for solving such combinatorial problems. The intent of the paper is to show how the algorithmic tools we present can be used as building blocks for higher level algorithms, and to present pointers to the literature for the reader to look up the specifics of these algorithms. We make no claim to completeness; a number of techniques have been omitted for brevity or because their chief application is not combinatorial in nature. In particular we give very little attention to parallel sorting, although sorting is used as a subroutine in a number of the algorithms we describe. We also only describe algorithms, and not lower bounds, for solving problems in parallel