Highly parallel computing of linear equations on the matrix-broadcast-memory connected array processor system
Yukio Kaneda, Masaki Kohata · 1982
The authors describe an efficient method for computing a set of n linear simultaneous equations in o(n) time, on the proposed array processor system. The proposed system is composed of a large number of microprocessors and its layout is p*q square. The microprocessors (hereafter called processing elements, or PEs) are connected to the row and columnwise broadcast memories. 2 references.