Modeling synthetic aperture radar computation with Aspen. (August 2013)

• Record Type: Journal Article
• Authors: Spafford, Kyle; Vetter, Jeffrey S.; Benson, Thomas; Parker, Mike
• Is Part Of: International journal of high performance computing applications. Volume 27:Number 3(2013)
• Page Start: 255
• Page End: 262
• Publication Date: 2013-08
• Subjects: performance modeling -- Aspen -- workload characterization -- Echelon -- synthetic aperture radar -- FFT; High performance computing -- Periodicals; Supercomputers -- Periodicals; 004.1105
• DOI: 10.1177/1094342013488262
• Languages: English
• ISSNs: 1094-3420

Abstract

This case study presents an analytical performance model for the DARPA UHPC streaming sensor challenge problem developed using Aspen, a domain-specific language for performance modeling. The model focuses on the exploration of algorithmic tradeoffs, data structures and storage, and the impact of an important tiling factor in the image formation kernel of a synthetic aperture radar image-processing computation.