Hypercubic storage layout and transforms in arbitrary dimensions using GPUs and CUDA

Hawick, K. A.; Playne, D. P.

Abstract

Many simulations in the physical sciences are expressed in terms of rectilinear arrays of variables. It is attractive to develop such simulations for use in 1-, 2-, 3- or arbitrary physical dimensions and also in a manner that supports exploitation of data-parallelism on fast modern processing devices. We report on data layouts and transformation algorithms that support both conventional and data-parallel memory layouts. We present our implementations expressed in both conventional serial C code as well as in NVIDIA's Compute Unified Device Architecture concurrent programming language for use on general purpose graphical processing units. We discuss: general memory layouts; specific optimizations possible for dimensions that are powers-of-two and common transformations, such as inverting, shifting and crinkling. We present performance data for some illustrative scientific applications of these layouts and transforms using several current GPU devices and discuss the code and speed scalability of this approach.

Citation

Hawick, K. A., & Playne, D. P. (2011). Hypercubic storage layout and transforms in arbitrary dimensions using GPUs and CUDA. Concurrency and Computation: Practice and Experience, 23(10), 1027-1050. https://doi.org/10.1002/cpe.1628

• Journal Article Type: Article
• Online Publication Date: Aug 28, 2010
• Publication Date: 2011-07
• Deposit Date: Nov 13, 2014
• Journal: Concurrency And Computation-Practice & Experience
• Print ISSN: 1532-0626
• Publisher: Wiley
• Peer Reviewed: Peer Reviewed
• Volume: 23
• Issue: 10
• Pages: 1027-1050
• DOI: https://doi.org/10.1002/cpe.1628
• Keywords: Theoretical Computer Science; Computer Networks and Communications; Computational Theory and Mathematics; Software; Computer Science Applications
• Public URL: https://hull-repository.worktribe.com/output/470690
• Contract Date: Nov 13, 2014