Autotuning Wavefront Abstractions for Heterogeneous Architectures
Institute for Computing Systems Architecture, University of Edinburgh, 10 Crichton Street, Edinburgh, UK
2012 Third Workshop on Applications for Multi-Core Architectures (WAMCA ’12), 2012
@article{mohanty2012autotuning,
title={Autotuning Wavefront Abstractions for Heterogeneous Architectures},
author={Mohanty, S. and Cole, M.},
year={2012}
}
We present our autotuned heterogeneous parallel programming abstraction for the wavefront 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 wavefront examples.
September 21, 2012 by hgpu
