Splotch: porting and optimizing for the Xeon Phi
School of Creative Technologies, University of Portsmouth, Portsmouth, UK
arXiv:1606.04427 [cs.DC], (14 Jun 2016)
@article{dykes2016splotch,
title={Splotch: porting and optimizing for the Xeon Phi},
author={Dykes, Timothy and Gheller, Claudio and Rivi, Marzia and Krokos, Mel},
year={2016},
month={jun},
archivePrefix={"arXiv"},
primaryClass={cs.DC}
}
With the increasing size and complexity of data produced by large scale numerical simulations, it is of primary importance for scientists to be able to exploit all available hardware in heterogenous High Performance Computing environments for increased throughput and efficiency. We focus on the porting and optimization of Splotch, a scalable visualization algorithm, to utilize the Xeon Phi, Intel’s coprocessor based upon the new Many Integrated Core architecture. We discuss steps taken to offload data to the coprocessor and algorithmic modifications to aid faster processing on the many-core architecture and make use of the uniquely wide vector capabilities of the device, with accompanying performance results using multiple Xeon Phi. Finally performance is compared against results achieved with the GPU implementation of Splotch.
June 16, 2016 by hgpu
