Publication Details
Evaluation of the Suitability of Intel Xeon Phi Clusters for the Simulation of Ultrasound Wave Propagation using Pseudospectral Methods
ltrasound simulations, Pseudospectral methods, k-Wave toolbox, Intel Xeon Phi, KNC, KNL, MPI, OpenMP, performance evaluation, scaling
Large-scale ultrasound simulations has excited high interest in medical applications such as high intensity focused ultrasound treatment or photoacoustic imaging. However, routine execution of such simulations using modern pseudospectral methods is computationally very challenging. To enable fast simulation, a cluster of computers is typically used. Nowadays, the trend in parallel computing is towards the use of accelerated nodes where the hard work is offloaded from processors to accelerators. During last five years, Intel has released two generations of accelerators called Intel Xeon Phi. The goal of this paper is to investigate the parameters on both architectures with respect to current processors, and evaluate the suitability of accelerated clusters for the distributed simulation of ultrasound propagation in medical applications. The paper reveals that the former version of Xeon Phis, the Knight's Corner architecture, suffers from several flaws that reduces the performance far below the Haswell processors. On the other hand, the second generation called Knight's Landing shows a very promising performance comparable with current processors.
@inproceedings{BUT158072,
author="Filip {Vaverka} and Bradley {Treeby} and Jiří {Jaroš}",
title="Evaluation of the Suitability of Intel Xeon Phi Clusters for the Simulation of Ultrasound Wave Propagation using Pseudospectral Methods",
booktitle="19th International Conference on Computational Science 2019, Faro, Portugal, June 12-14, 2019",
year="2019",
series="Lecture Notes in Computer Science",
volume="11538",
pages="577--590",
publisher="Springer International Publishing",
address="Faro",
doi="10.1007/978-3-030-22744-9\{_}45",
isbn="978-3-030-22743-2",
url="https://link.springer.com/chapter/10.1007/978-3-030-22744-9_45"
}