Please use this identifier to cite or link to this item:
Title: Preconditioned GMRES solver on multiple GPU architecture
Authors: Chen, Zhangxing (John)
Yang, B.
Liu, Hui
Keywords: Mathematics;Computer Science;Engineering-Petroleum
Issue Date: Aug-2016
Publisher: Computers and Mathematics with Applications
Series/Report no.: 72;1076-1095
Abstract: In this paper, we analyze the preconditioned GMRES algorithm in detail and decompose it into components to implement on multiple-GPU architecture. The operations of vector updates, dot products and Sparse Matrix-Vector multiplication (SpMV) are implemented in parallel. In addition, a specific communication mechanism for SpMV is designed. The preconditioner is established on the host (CPU) and solved on the devices (GPUs). Validated by a series of numerical experiments, the GPU-based GMRES solver is effective and favorable parallel performance is achieved.
Appears in Collections:Chen, Zhangxing (John)

Files in This Item:
File Description SizeFormat 
ucalgary_2015_yang_bo.pdfElectronic Theses973.85 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.