A compact algorithm for Gaussian elimination over GF(2) implemented on highly parallel computers (Q761009)

Gaussian elimination over GF(2) is an essential part of the factorization routine of \textit{M. A. Morrison} and \textit{J. Brillhart} [Math. Comput. 29, 183-205 (1975; Zbl 0302.10010)]. There it is required to find those subsets of a given set of integers whose product is a perfect square. It is also desirable to preserve the history matrix so that the subset is available after reduction, without requiring extra space. The authors report on programs for an ICL-DAP parallel array processor viewed as 4096 distinct bit processors each with a memory of 16384 bits. The number of processors used corresponds to the number of matrix columns, for example, with rows handled in parallel. [Further details are referred to the second author, Numerical mathematics and computing, Proc. 12th Manitoba Conf., Winnepeg/Manit. 1982, Congr. Numerantium 38, 261-270 (1983; Zbl 0536.68024)].