An Algorithm-by-Blocks for SuperMatrix Band Cholesky Factorization [chapter]

Gregorio Quintana-Ortí, Enrique S. Quintana-Ortí, Alfredo Remón, Robert A. van de Geijn
2008 Lecture Notes in Computer Science  
We pursue the scalable parallel implementation of the factorization of band matrices with medium to large bandwidth targeting SMP and multi-core architectures. Our approach decomposes the computation into a large number of fine-grained operations exposing a higher degree of parallelism. The SuperMatrix run-time system allows an out-of-order scheduling of operations that is transparent to the programmer. Experimental results for the Cholesky factorization of band matrices on two parallel
more » ... s with sixteen processors demonstrate the scalability of the solution.
doi:10.1007/978-3-540-92859-1_21 fatcat:4jv4sopocncjpm3a665pn3jjh4