AN EFFICIENT BLOCK-ORIENTED APPROACH TO PARALLEL SPARSE CHOLESKY FACTORIZATION

This paper explores the use of a subblock decomposition strategy for p arallel sparse Cholesky factorization in which the sparse matrix is de composed into rectangular blocks. Such a strategy has enormous theoret ical scalability advantages over more traditional column-oriented and panel-oriented decompositions. However, little progress has been made in producing a practical subblock method. This paper describes and eva luates an approach that is simple to implement, provides slightly high er performance than column (and panel) methods on small parallel machi nes, and has the potential to provide much higher performance on large parallel machines.