Accelerating Multi-Process Communication for Parallel 3-D FFT

Alan Ayala; Stan Tomov; Miroslav Stoyanov; Azzam Haidar; Jack Dongarra

doi:10.1109/ExaMPI54564.2021.00011

Accelerating Multi-Process Communication for Parallel 3-D FFT

Alan Ayala, Stan Tomov, Miroslav Stoyanov, Azzam Haidar, Jack Dongarra

Multiscale Methods

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

4 Scopus citations

Abstract

Today largest and most powerful supercomputers in the world are built on heterogeneous platforms; and using the combined power of multi-core CPUs and GPUs, has had a great impact accelerating large-scale applications. However, on these architectures, parallel algorithms, such as the Fast Fourier Transform (FFT), encounter that inter-processor communication become a bottleneck and limits their scalability. In this paper, we present techniques for speeding up multi-process communication cost during the computation of FFTs, considering hybrid network connections as those expected on upcoming exascale machines. Among our techniques, we present algorithmic tuning, making use of phase diagrams; parametric tuning, using different FFT settings; and MPI distribution tuning based on FFT size and computational resources available. We present several experiments obtained on Summit supercomputer at Oak Ridge National Laboratory, using up to 40,960 IBM Power9 cores and 6,144 NVIDIA V-100 GPUs.

Original language	English
Title of host publication	Proceedings of ExaMPI 2021
Subtitle of host publication	Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	46-53
Number of pages	8
ISBN (Electronic)	9781665411080
DOIs	https://doi.org/10.1109/ExaMPI54564.2021.00011
State	Published - 2021
Event	2021 Workshop on Exascale MPI, ExaMPI 2021 - St. Louis, United States Duration: Nov 14 2021 → …

Publication series

Name	Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis

Conference

Conference	2021 Workshop on Exascale MPI, ExaMPI 2021
Country/Territory	United States
City	St. Louis
Period	11/14/21 → …

Funding

ACKNOWLEDGMENT This research was supported by the Exascale Computing Project (ECP), Project Number: 17-SC-20-SC, a collaborative effort of two DOE organizations (the Office of Science and the National Nuclear Security Administration) responsible for the planning and preparation of a capable exascale ecosystem.

Keywords

Exascale FFT
Hybrid systems
MPI tuning
Scalability

Access to Document

10.1109/ExaMPI54564.2021.00011

Cite this

Ayala, A., Tomov, S., Stoyanov, M., Haidar, A., & Dongarra, J. (2021). Accelerating Multi-Process Communication for Parallel 3-D FFT. In Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 46-53). (Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ExaMPI54564.2021.00011

Ayala, Alan ; Tomov, Stan ; Stoyanov, Miroslav et al. / Accelerating Multi-Process Communication for Parallel 3-D FFT. Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 46-53 (Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis).

@inproceedings{0728307ad37c4fbc91fd2640e272b332,

title = "Accelerating Multi-Process Communication for Parallel 3-D FFT",

abstract = "Today largest and most powerful supercomputers in the world are built on heterogeneous platforms; and using the combined power of multi-core CPUs and GPUs, has had a great impact accelerating large-scale applications. However, on these architectures, parallel algorithms, such as the Fast Fourier Transform (FFT), encounter that inter-processor communication become a bottleneck and limits their scalability. In this paper, we present techniques for speeding up multi-process communication cost during the computation of FFTs, considering hybrid network connections as those expected on upcoming exascale machines. Among our techniques, we present algorithmic tuning, making use of phase diagrams; parametric tuning, using different FFT settings; and MPI distribution tuning based on FFT size and computational resources available. We present several experiments obtained on Summit supercomputer at Oak Ridge National Laboratory, using up to 40,960 IBM Power9 cores and 6,144 NVIDIA V-100 GPUs.",

keywords = "Exascale FFT, Hybrid systems, MPI tuning, Scalability",

author = "Alan Ayala and Stan Tomov and Miroslav Stoyanov and Azzam Haidar and Jack Dongarra",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 Workshop on Exascale MPI, ExaMPI 2021 ; Conference date: 14-11-2021",

year = "2021",

doi = "10.1109/ExaMPI54564.2021.00011",

language = "English",

series = "Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis",

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

pages = "46--53",

booktitle = "Proceedings of ExaMPI 2021",

}

Ayala, A, Tomov, S, Stoyanov, M, Haidar, A & Dongarra, J 2021, Accelerating Multi-Process Communication for Parallel 3-D FFT. in Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis. Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis, Institute of Electrical and Electronics Engineers Inc., pp. 46-53, 2021 Workshop on Exascale MPI, ExaMPI 2021, St. Louis, United States, 11/14/21. https://doi.org/10.1109/ExaMPI54564.2021.00011

Accelerating Multi-Process Communication for Parallel 3-D FFT. / Ayala, Alan; Tomov, Stan; Stoyanov, Miroslav et al.
Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2021. p. 46-53 (Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis).

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

TY - GEN

T1 - Accelerating Multi-Process Communication for Parallel 3-D FFT

AU - Ayala, Alan

AU - Tomov, Stan

AU - Stoyanov, Miroslav

AU - Haidar, Azzam

AU - Dongarra, Jack

PY - 2021

Y1 - 2021

N2 - Today largest and most powerful supercomputers in the world are built on heterogeneous platforms; and using the combined power of multi-core CPUs and GPUs, has had a great impact accelerating large-scale applications. However, on these architectures, parallel algorithms, such as the Fast Fourier Transform (FFT), encounter that inter-processor communication become a bottleneck and limits their scalability. In this paper, we present techniques for speeding up multi-process communication cost during the computation of FFTs, considering hybrid network connections as those expected on upcoming exascale machines. Among our techniques, we present algorithmic tuning, making use of phase diagrams; parametric tuning, using different FFT settings; and MPI distribution tuning based on FFT size and computational resources available. We present several experiments obtained on Summit supercomputer at Oak Ridge National Laboratory, using up to 40,960 IBM Power9 cores and 6,144 NVIDIA V-100 GPUs.

AB - Today largest and most powerful supercomputers in the world are built on heterogeneous platforms; and using the combined power of multi-core CPUs and GPUs, has had a great impact accelerating large-scale applications. However, on these architectures, parallel algorithms, such as the Fast Fourier Transform (FFT), encounter that inter-processor communication become a bottleneck and limits their scalability. In this paper, we present techniques for speeding up multi-process communication cost during the computation of FFTs, considering hybrid network connections as those expected on upcoming exascale machines. Among our techniques, we present algorithmic tuning, making use of phase diagrams; parametric tuning, using different FFT settings; and MPI distribution tuning based on FFT size and computational resources available. We present several experiments obtained on Summit supercomputer at Oak Ridge National Laboratory, using up to 40,960 IBM Power9 cores and 6,144 NVIDIA V-100 GPUs.

KW - Exascale FFT

KW - Hybrid systems

KW - MPI tuning

KW - Scalability

UR - http://www.scopus.com/inward/record.url?scp=85124654321&partnerID=8YFLogxK

U2 - 10.1109/ExaMPI54564.2021.00011

DO - 10.1109/ExaMPI54564.2021.00011

M3 - Conference contribution

AN - SCOPUS:85124654321

T3 - Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis

SP - 46

EP - 53

BT - Proceedings of ExaMPI 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2021 Workshop on Exascale MPI, ExaMPI 2021

Y2 - 14 November 2021

ER -

Ayala A, Tomov S, Stoyanov M, Haidar A, Dongarra J. Accelerating Multi-Process Communication for Parallel 3-D FFT. In Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc. 2021. p. 46-53. (Proceedings of ExaMPI 2021: Workshop on Exascale MPI, Held in conjunction with SC 2021: The International Conference for High Performance Computing, Networking, Storage and Analysis). doi: 10.1109/ExaMPI54564.2021.00011

Accelerating Multi-Process Communication for Parallel 3-D FFT

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this