Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor

Wei Jia He; Ming Lin Yang; Xin Qing Sheng

doi:10.1109/NEMO49486.2020.9343619

Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor

Wei Jia He, Ming Lin Yang, Xin Qing Sheng

School of Integrated Circuits and Electronics

Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

An efficient many-core parallel implementation of the multilevel fast multipole algorithm (MLFMA) for 3D electromagnetic scattering problems on SW26010 processor using the Athread programming model is presented. The performance achievements are investigated in terms of accuracy, speed up, and scalability. Numerical results demonstrate that our MLFMA implementation on SW26010 is capable for solving real-life problems with over one million unknowns with a remarkable speed-up(up to 8x).

Original language	English
Title of host publication	2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728169668
DOIs	https://doi.org/10.1109/NEMO49486.2020.9343619
Publication status	Published - 7 Dec 2020
Event	2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020 - Hangzhou, China Duration: 7 Dec 2020 → 9 Dec 2020

Publication series

Name	2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020

Conference

Conference	2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020
Country/Territory	China
City	Hangzhou
Period	7/12/20 → 9/12/20

Access to Document

10.1109/NEMO49486.2020.9343619

Cite this

He, W. J., Yang, M. L., & Sheng, X. Q. (2020). Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor. In 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020 Article 9343619 (2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/NEMO49486.2020.9343619

He, Wei Jia ; Yang, Ming Lin ; Sheng, Xin Qing. / Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor. 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020. Institute of Electrical and Electronics Engineers Inc., 2020. (2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020).

@inproceedings{bfb1165193cf491e9f8f4b77d104d678,

title = "Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor",

abstract = "An efficient many-core parallel implementation of the multilevel fast multipole algorithm (MLFMA) for 3D electromagnetic scattering problems on SW26010 processor using the Athread programming model is presented. The performance achievements are investigated in terms of accuracy, speed up, and scalability. Numerical results demonstrate that our MLFMA implementation on SW26010 is capable for solving real-life problems with over one million unknowns with a remarkable speed-up(up to 8x).",

author = "He, {Wei Jia} and Yang, {Ming Lin} and Sheng, {Xin Qing}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020 ; Conference date: 07-12-2020 Through 09-12-2020",

year = "2020",

month = dec,

day = "7",

doi = "10.1109/NEMO49486.2020.9343619",

language = "English",

series = "2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020",

address = "United States",

}

He, WJ, Yang, ML & Sheng, XQ 2020, Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor. in 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020., 9343619, 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020, Institute of Electrical and Electronics Engineers Inc., 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020, Hangzhou, China, 7/12/20. https://doi.org/10.1109/NEMO49486.2020.9343619

Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor. / He, Wei Jia; Yang, Ming Lin; Sheng, Xin Qing.
2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020. Institute of Electrical and Electronics Engineers Inc., 2020. 9343619 (2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor

AU - He, Wei Jia

AU - Yang, Ming Lin

AU - Sheng, Xin Qing

PY - 2020/12/7

Y1 - 2020/12/7

N2 - An efficient many-core parallel implementation of the multilevel fast multipole algorithm (MLFMA) for 3D electromagnetic scattering problems on SW26010 processor using the Athread programming model is presented. The performance achievements are investigated in terms of accuracy, speed up, and scalability. Numerical results demonstrate that our MLFMA implementation on SW26010 is capable for solving real-life problems with over one million unknowns with a remarkable speed-up(up to 8x).

AB - An efficient many-core parallel implementation of the multilevel fast multipole algorithm (MLFMA) for 3D electromagnetic scattering problems on SW26010 processor using the Athread programming model is presented. The performance achievements are investigated in terms of accuracy, speed up, and scalability. Numerical results demonstrate that our MLFMA implementation on SW26010 is capable for solving real-life problems with over one million unknowns with a remarkable speed-up(up to 8x).

UR - http://www.scopus.com/inward/record.url?scp=85101193424&partnerID=8YFLogxK

U2 - 10.1109/NEMO49486.2020.9343619

DO - 10.1109/NEMO49486.2020.9343619

M3 - Conference contribution

AN - SCOPUS:85101193424

T3 - 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020

BT - 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020

Y2 - 7 December 2020 through 9 December 2020

ER -

He WJ, Yang ML, Sheng XQ. Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor. In 2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020. Institute of Electrical and Electronics Engineers Inc. 2020. 9343619. (2020 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2020). doi: 10.1109/NEMO49486.2020.9343619

Efficient Implementation of Multilevel Fast Multipole Algorithm on SW26010 Many-core Processor

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this