Full Text Views
19
In this paper, a Many Integrated Core Architecture (MIC) accelerated parallel method of moment (MoM) algorithm is proposed to solve electromagnetic problems in practical applications, where MIC means a kind of coprocessor or accelerator in computer systems which is used to accelerate the computation performed by Central Processing Unit (CPU). Three critical points are introduced in this paper in detail. The first one is the design of the parallel framework, which ensures that the algorithm can run on distributed memory platform with multiple nodes. The hybrid Message Passing Interface (MPI) and Open Multi-Processing (OpenMP) programming model is designed to achieve the purposes. The second one is the out-of-core algorithm, which greatly breaks the restriction of MIC memory. The third one is the pipeline algorithm which overlaps the data movement with MIC computation. The pipeline algorithm successfully hides the communication and thus greatly enhances the performance of hybrid MIC/CPU MoM. Numerical result indicates that the proposed algorithm has good parallel efficiency and scalability, and twice faster performance when compared with the corresponding CPU algorithm.
Yan CHEN
Xidian University
Yu ZHANG
Xidian University
Guanghui ZHANG
Xidian University
Xunwang ZHAO
Xidian University
ShaoHua WU
Inspur
Qing ZHANG
Inspur
XiaoPeng YANG
Beijing Institute of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yan CHEN, Yu ZHANG, Guanghui ZHANG, Xunwang ZHAO, ShaoHua WU, Qing ZHANG, XiaoPeng YANG, "Hybrid MIC/CPU Parallel Implementation of MoM on MIC Cluster for Electromagnetic Problems" in IEICE TRANSACTIONS on Electronics,
vol. E99-C, no. 7, pp. 735-743, July 2016, doi: 10.1587/transele.E99.C.735.
Abstract: In this paper, a Many Integrated Core Architecture (MIC) accelerated parallel method of moment (MoM) algorithm is proposed to solve electromagnetic problems in practical applications, where MIC means a kind of coprocessor or accelerator in computer systems which is used to accelerate the computation performed by Central Processing Unit (CPU). Three critical points are introduced in this paper in detail. The first one is the design of the parallel framework, which ensures that the algorithm can run on distributed memory platform with multiple nodes. The hybrid Message Passing Interface (MPI) and Open Multi-Processing (OpenMP) programming model is designed to achieve the purposes. The second one is the out-of-core algorithm, which greatly breaks the restriction of MIC memory. The third one is the pipeline algorithm which overlaps the data movement with MIC computation. The pipeline algorithm successfully hides the communication and thus greatly enhances the performance of hybrid MIC/CPU MoM. Numerical result indicates that the proposed algorithm has good parallel efficiency and scalability, and twice faster performance when compared with the corresponding CPU algorithm.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/transele.E99.C.735/_p
Copy
@ARTICLE{e99-c_7_735,
author={Yan CHEN, Yu ZHANG, Guanghui ZHANG, Xunwang ZHAO, ShaoHua WU, Qing ZHANG, XiaoPeng YANG, },
journal={IEICE TRANSACTIONS on Electronics},
title={Hybrid MIC/CPU Parallel Implementation of MoM on MIC Cluster for Electromagnetic Problems},
year={2016},
volume={E99-C},
number={7},
pages={735-743},
abstract={In this paper, a Many Integrated Core Architecture (MIC) accelerated parallel method of moment (MoM) algorithm is proposed to solve electromagnetic problems in practical applications, where MIC means a kind of coprocessor or accelerator in computer systems which is used to accelerate the computation performed by Central Processing Unit (CPU). Three critical points are introduced in this paper in detail. The first one is the design of the parallel framework, which ensures that the algorithm can run on distributed memory platform with multiple nodes. The hybrid Message Passing Interface (MPI) and Open Multi-Processing (OpenMP) programming model is designed to achieve the purposes. The second one is the out-of-core algorithm, which greatly breaks the restriction of MIC memory. The third one is the pipeline algorithm which overlaps the data movement with MIC computation. The pipeline algorithm successfully hides the communication and thus greatly enhances the performance of hybrid MIC/CPU MoM. Numerical result indicates that the proposed algorithm has good parallel efficiency and scalability, and twice faster performance when compared with the corresponding CPU algorithm.},
keywords={},
doi={10.1587/transele.E99.C.735},
ISSN={1745-1353},
month={July},}
Copy
TY - JOUR
TI - Hybrid MIC/CPU Parallel Implementation of MoM on MIC Cluster for Electromagnetic Problems
T2 - IEICE TRANSACTIONS on Electronics
SP - 735
EP - 743
AU - Yan CHEN
AU - Yu ZHANG
AU - Guanghui ZHANG
AU - Xunwang ZHAO
AU - ShaoHua WU
AU - Qing ZHANG
AU - XiaoPeng YANG
PY - 2016
DO - 10.1587/transele.E99.C.735
JO - IEICE TRANSACTIONS on Electronics
SN - 1745-1353
VL - E99-C
IS - 7
JA - IEICE TRANSACTIONS on Electronics
Y1 - July 2016
AB - In this paper, a Many Integrated Core Architecture (MIC) accelerated parallel method of moment (MoM) algorithm is proposed to solve electromagnetic problems in practical applications, where MIC means a kind of coprocessor or accelerator in computer systems which is used to accelerate the computation performed by Central Processing Unit (CPU). Three critical points are introduced in this paper in detail. The first one is the design of the parallel framework, which ensures that the algorithm can run on distributed memory platform with multiple nodes. The hybrid Message Passing Interface (MPI) and Open Multi-Processing (OpenMP) programming model is designed to achieve the purposes. The second one is the out-of-core algorithm, which greatly breaks the restriction of MIC memory. The third one is the pipeline algorithm which overlaps the data movement with MIC computation. The pipeline algorithm successfully hides the communication and thus greatly enhances the performance of hybrid MIC/CPU MoM. Numerical result indicates that the proposed algorithm has good parallel efficiency and scalability, and twice faster performance when compared with the corresponding CPU algorithm.
ER -