MPC Related Computational Capabilities of ARMv7A Processors

Gianluca Frison, John Bagterp Jørgensen

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

In recent years, the mass market of mobile devices has pushed the demand for increasingly fast but cheap processors. ARM, the world leader in this sector, has developed the Cortex-A series of processors with focus on computationally intensive applications. If properly programmed, these processors are powerful enough to solve the complex optimization problems arising in MPC in real-time, while keeping the traditional low-cost and low-power consumption. This makes these processors ideal candidates for use in embedded MPC. In this paper, we investigate the floating-point capabilities of Cortex A7, A9 and A15 and show how to exploit the unique features of each processor to obtain the best performance, in the context of a novel implementation method for the linear-algebra routines used in MPC solvers. This method adapts high-performance computing techniques to the needs of embedded MPC. In particular, we investigate the performance of matrix-matrix and matrix-vector multiplications, which are the backbones of second- and first-order methods for convex optimization. Finally, we test the performance of MPC solvers implemented using these optimized linear-algebra routines.
Original languageEnglish
Title of host publicationProceedings of the European Control Conference (ECC 2015)
PublisherIEEE
Publication date2015
Pages3414-3421
ISBN (Print)978-3-9524269-3-7
DOIs
Publication statusPublished - 2015
Event14th European Control Conference (ECC 2015) - Johannes Kepler University, Linz, Austria
Duration: 15 Jul 201517 Jul 2015
Conference number: 15
http://www.ecc15.at/

Conference

Conference14th European Control Conference (ECC 2015)
Number15
LocationJohannes Kepler University
CountryAustria
CityLinz
Period15/07/201517/07/2015
Internet address

Cite this

Frison, G., & Jørgensen, J. B. (2015). MPC Related Computational Capabilities of ARMv7A Processors. In Proceedings of the European Control Conference (ECC 2015) (pp. 3414-3421). IEEE. https://doi.org/10.1109/ECC.2015.7331062
Frison, Gianluca ; Jørgensen, John Bagterp. / MPC Related Computational Capabilities of ARMv7A Processors. Proceedings of the European Control Conference (ECC 2015). IEEE, 2015. pp. 3414-3421
@inproceedings{1744a3c2646b42de8fb8744f898a7398,
title = "MPC Related Computational Capabilities of ARMv7A Processors",
abstract = "In recent years, the mass market of mobile devices has pushed the demand for increasingly fast but cheap processors. ARM, the world leader in this sector, has developed the Cortex-A series of processors with focus on computationally intensive applications. If properly programmed, these processors are powerful enough to solve the complex optimization problems arising in MPC in real-time, while keeping the traditional low-cost and low-power consumption. This makes these processors ideal candidates for use in embedded MPC. In this paper, we investigate the floating-point capabilities of Cortex A7, A9 and A15 and show how to exploit the unique features of each processor to obtain the best performance, in the context of a novel implementation method for the linear-algebra routines used in MPC solvers. This method adapts high-performance computing techniques to the needs of embedded MPC. In particular, we investigate the performance of matrix-matrix and matrix-vector multiplications, which are the backbones of second- and first-order methods for convex optimization. Finally, we test the performance of MPC solvers implemented using these optimized linear-algebra routines.",
author = "Gianluca Frison and J{\o}rgensen, {John Bagterp}",
year = "2015",
doi = "10.1109/ECC.2015.7331062",
language = "English",
isbn = "978-3-9524269-3-7",
pages = "3414--3421",
booktitle = "Proceedings of the European Control Conference (ECC 2015)",
publisher = "IEEE",
address = "United States",

}

Frison, G & Jørgensen, JB 2015, MPC Related Computational Capabilities of ARMv7A Processors. in Proceedings of the European Control Conference (ECC 2015). IEEE, pp. 3414-3421, 14th European Control Conference (ECC 2015), Linz, Austria, 15/07/2015. https://doi.org/10.1109/ECC.2015.7331062

MPC Related Computational Capabilities of ARMv7A Processors. / Frison, Gianluca; Jørgensen, John Bagterp.

Proceedings of the European Control Conference (ECC 2015). IEEE, 2015. p. 3414-3421.

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - MPC Related Computational Capabilities of ARMv7A Processors

AU - Frison, Gianluca

AU - Jørgensen, John Bagterp

PY - 2015

Y1 - 2015

N2 - In recent years, the mass market of mobile devices has pushed the demand for increasingly fast but cheap processors. ARM, the world leader in this sector, has developed the Cortex-A series of processors with focus on computationally intensive applications. If properly programmed, these processors are powerful enough to solve the complex optimization problems arising in MPC in real-time, while keeping the traditional low-cost and low-power consumption. This makes these processors ideal candidates for use in embedded MPC. In this paper, we investigate the floating-point capabilities of Cortex A7, A9 and A15 and show how to exploit the unique features of each processor to obtain the best performance, in the context of a novel implementation method for the linear-algebra routines used in MPC solvers. This method adapts high-performance computing techniques to the needs of embedded MPC. In particular, we investigate the performance of matrix-matrix and matrix-vector multiplications, which are the backbones of second- and first-order methods for convex optimization. Finally, we test the performance of MPC solvers implemented using these optimized linear-algebra routines.

AB - In recent years, the mass market of mobile devices has pushed the demand for increasingly fast but cheap processors. ARM, the world leader in this sector, has developed the Cortex-A series of processors with focus on computationally intensive applications. If properly programmed, these processors are powerful enough to solve the complex optimization problems arising in MPC in real-time, while keeping the traditional low-cost and low-power consumption. This makes these processors ideal candidates for use in embedded MPC. In this paper, we investigate the floating-point capabilities of Cortex A7, A9 and A15 and show how to exploit the unique features of each processor to obtain the best performance, in the context of a novel implementation method for the linear-algebra routines used in MPC solvers. This method adapts high-performance computing techniques to the needs of embedded MPC. In particular, we investigate the performance of matrix-matrix and matrix-vector multiplications, which are the backbones of second- and first-order methods for convex optimization. Finally, we test the performance of MPC solvers implemented using these optimized linear-algebra routines.

U2 - 10.1109/ECC.2015.7331062

DO - 10.1109/ECC.2015.7331062

M3 - Article in proceedings

SN - 978-3-9524269-3-7

SP - 3414

EP - 3421

BT - Proceedings of the European Control Conference (ECC 2015)

PB - IEEE

ER -

Frison G, Jørgensen JB. MPC Related Computational Capabilities of ARMv7A Processors. In Proceedings of the European Control Conference (ECC 2015). IEEE. 2015. p. 3414-3421 https://doi.org/10.1109/ECC.2015.7331062