Publications Details
Parallel QR factorization on a hypercube using the torus wrap mapping
We present an algorithm for the QR factorization of a dense matrix without column pivoting on a hypercube multiprocessor. The algorithm combines the optimal numerical efficiency of Householder reflections with the excellent communication properties of the torus wrap mapping. Analytical results indicate that the communication cost for this algorithm is less than that for other common approaches. Numerical results on an nCUBE 2 confirm the efficiency of our technique. 23 refs., 5 figs., 1 tab.