DocumentCode
2422562
Title
Autotuning Wavefront Abstractions for Heterogeneous Architectures
Author
Mohanty, Siddharth ; Cole, Murray
Author_Institution
Inst. for Comput. Syst. Archit., Univ. of Edinburgh, Edinburgh, UK
fYear
2012
fDate
24-25 Oct. 2012
Firstpage
42
Lastpage
47
Abstract
We present our auto tuned heterogeneous parallel programming abstraction for the wave front pattern. An exhaustive search of the tuning space indicates that correct setting of tuning factors can average 37x speedup over a sequential baseline. Our best automated machine learning based heuristic obtains 92% of this ideal speedup, averaged across our full range of wave front examples.
Keywords
learning (artificial intelligence); parallel programming; automated machine learning; autotuning wavefront abstractions; heterogeneous architectures; parallel programming; wave front pattern; Graphics processing units; Kernel; Parallel processing; Support vector machines; Tiles; Tuners; Abstractions; Autotuning; GPU; Heterogeneous Computing; Wavefront;
fLanguage
English
Publisher
ieee
Conference_Titel
Applications for Multi-Core Architectures (WAMCA), 2012 Third Workshop on
Conference_Location
New York, NY
Print_ISBN
978-1-4673-5025-9
Type
conf
DOI
10.1109/WAMCA.2012.14
Filename
6374751
Link To Document