Real-Time Dedispersion for Fast Radio Transient Surveys, using Auto Tuning on Many-Core Accelerators
Faculty of Sciences, Vrije Universiteit Amsterdam, Amsterdam, the Netherlands
arXiv:1601.01165 [astro-ph.IM], (6 Jan 2016)
@article{sclocco2016realtime,
title={Real-Time Dedispersion for Fast Radio Transient Surveys, using Auto Tuning on Many-Core Accelerators},
author={Sclocco, Alessio and Leeuwen, Joeri van and Bal, Henri E. and Nieuwpoort, Rob V. van},
year={2016},
month={jan},
archivePrefix={"arXiv"},
primaryClass={astro-ph.IM},
doi={10.1016/j.ascom.2016.01.001}
}
Dedispersion, the removal of deleterious smearing of impulsive signals by the interstellar matter, is one of the most intensive processing steps in any radio survey for pulsars and fast transients. We here present a study of the parallelization of this algorithm on many-core accelerators, including GPUs from AMD and NVIDIA, and the Intel Xeon Phi. We find that dedispersion is inherently memory-bound. Even in a perfect scenario, hardware limitations keep the arithmetic intensity low, thus limiting performance. We next exploit auto-tuning to adapt dedispersion to different accelerators, observations, and even telescopes. We demonstrate that the optimal settings differ between observational setups, and that auto-tuning significantly improves performance. This impacts time-domain surveys from Apertif to SKA.
January 12, 2016 by hgpu
