E. Ng et Bw. Peyton, A SUPERNODAL CHOLESKY FACTORIZATION ALGORITHM FOR SHARED-MEMORY MULTIPROCESSORS, SIAM journal on scientific computing, 14(4), 1993, pp. 761-769
This paper presents a parallel sparse Cholesky factorization algorithm
for shared-memory MIMD multiprocessors. The algorithm is particularly
well suited for vector supercomputers with multiple processors, such
as the Cray Y-MR The new algorithm is a straightforward parallelizatio
n of the left-looking supernodal sparse Cholesky factorization algorit
hm. Like its sequential predecessor, it improves performance by reduci
ng indirect addressing and memory traffic. Experimental results on a C
ray Y-MP demonstrate the effectiveness of the new algorithm. On eight
processors of a Cray Y-MP, the new routine performs the factorization
at rates exceeding one Gflop for several test problems from the Harwel
l-Boeing sparse matrix collection.