EECS Publication
Fast Cholesky Factorization on GPUs for Batch and Native Modes in MAGMA
Ahmad Abdelfatah and Azzam Haidar and Stanimire Tomov and Jack Dongarra
This paper presents a GPU-accelerated Cholesky factorization for two different modes of operation. The first one is the batch mode, where many independent factorizations on small matrices can be performed concurrently. This mode supports fixed size and variable size problems, and is found in many scientific applications. The second mode is the native mode, where one factorization is performed on a large matrix without any CPU involvement, which allows the CPU do other useful work. We show that, despite the different workloads, both modes of operation share a common code-base that uses the GPU only. We also show that the developed routines achieve significant speedups against a multi-core CPU using the MKL library. This work is part of the MAGMA library. Keywords: GPU computing, Cholesky factorization, batched execution
Published 2017-01-04 05:00:00 as ut-eecs-16-748 (ID:608)