Comparing Hybrid CPU-GPU and Native GPU-only Acceleration for Linear Algebra

TitleComparing Hybrid CPU-GPU and Native GPU-only Acceleration for Linear Algebra
Publication TypeConference Paper
Year of Publication2015
AuthorsGates, M., S. Tomov, and A. Haidar
Conference Name2015 SIAM Conference on Applied Linear Algebra
Date Published10-2015
PublisherSIAM
Conference LocationAtlanta, GA
AbstractAccelerating dense linear algebra using GPUs admits two models: hybrid CPU-GPU and GPU-only. The hybrid model factors the panel on the CPU while updating the trailing matrix on the GPU, concentrating the GPU on high-performance matrix multiplies. The GPU-only model performs the entire computation on the GPU, avoiding costly data transfers to the CPU. We compare these two approaches for three QR-based algorithms: QR factorization, rank revealing QR, and reduction to Hessenberg.
Project Tags: