Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs. (March 2018)

• Record Type: Journal Article
• Authors: Anzt, Hartwig; Kreutzer, Moritz; Ponce, Eduardo; Peterson, Gregory D; Wellein, Gerhard; Dongarra, Jack
• Is Part Of: International journal of high performance computing applications. Volume 32:Number 2(2018)
• Page Start: 220
• Page End: 230
• Publication Date: 2018-03
• Subjects: Induced dimension reduction (IDR) -- GPU -- co-design -- kernel fusion -- kernel overlap -- roofline performance model; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/1094342016646844
• Languages: English
• ISSNs: 1094-3420

Abstract

In this paper, we present an optimized GPU implementation for the induced dimension reduction algorithm. We improve data locality, combine it with an efficient sparse matrix vector kernel, and investigate the potential of overlapping computation with communication as well as the possibility of concurrent kernel execution. A comprehensive performance evaluation is conducted using a suitable performance model. The analysis reveals efficiency of up to 90%, which indicates that the implementation achieves performance close to the theoretically attainable bound.