A Multithreaded Algorithm for Sparse Cholesky Factorization on Hybrid Multicore Architectures

https://doi.org/10.1016/j.procs.2017.05.260Get rights and content
Under a Creative Commons license
open access

Abstract

We present a multithreaded method for supernodal sparse Cholesky factorization on a hybrid multicore platform consisting of a multicore CPU and GPU. Our algorithm can utilize concurrency at different levels of the elimination tree by using multiple threads in both the CPU and the GPU. The elimination tree is a tree data structure describing the workflow of the factorization. Our experiments results on a platform consisting of an Intel multicore processor along with an Nvidia GPU indicate a significant improvement in performance and energy over single-threaded supernodal algorithm.

Keywords

sparse matrices
sparse direct methods
Cholesky factorization
GPU
CUDA

Cited by (0)