Performance optimization of SpMV using CRS format by considering OpenMP scheduling on CPUs and MIC

Satoshi Ohshima, Takahiro Katagiri, Masaharu Matsumoto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

In this study, we evaluate the performance of sparse matrix-vector multiplication (SpMV) using the compressed row storage (CRS) format on CPUs and MIC. We focus on the relationship between OpenMP scheduling and performance. The performance of SpMV is measured using various OpenMP scheduling settings and the results are analyzed, which show that OpenMP scheduling has a considerable effect on the performance of SpMV. We confirm that some scheduling settings resulted in performance improvements compared with default scheduling for particular matrices. The results of the evaluation show that the performance of SpMV is improved by up to 1.57 times compared with SPARC64 IXfx, 2.47 times compared with Xeon Ivy Bridge-EP, and 2.26 times compared with Knights Corner. Next, we modify the SpMV function of OpenATLib, an auto-tuned numerical library, to consider the scheduling of optimization as an additional SpMV implementation. We measure the performance of the GMRES solver and obtain performance improvements of up to 11.4%. These results will help to improve the performance of various numerical calculation applications.

Original languageEnglish
Title of host publicationProceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages253-260
Number of pages8
ISBN (Electronic)9781479943050
DOIs
Publication statusPublished - Nov 6 2014
Event2014 8th IEEE International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014 - Aizu-Wakamatsu, Japan
Duration: Sep 23 2014Sep 25 2014

Publication series

NameProceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014

Other

Other2014 8th IEEE International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014
CountryJapan
CityAizu-Wakamatsu
Period9/23/149/25/14

Fingerprint

Program processors
Scheduling

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

Ohshima, S., Katagiri, T., & Matsumoto, M. (2014). Performance optimization of SpMV using CRS format by considering OpenMP scheduling on CPUs and MIC. In Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014 (pp. 253-260). [6949479] (Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/MCSoC.2014.43

Performance optimization of SpMV using CRS format by considering OpenMP scheduling on CPUs and MIC. / Ohshima, Satoshi; Katagiri, Takahiro; Matsumoto, Masaharu.

Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 253-260 6949479 (Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ohshima, S, Katagiri, T & Matsumoto, M 2014, Performance optimization of SpMV using CRS format by considering OpenMP scheduling on CPUs and MIC. in Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014., 6949479, Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014, Institute of Electrical and Electronics Engineers Inc., pp. 253-260, 2014 8th IEEE International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014, Aizu-Wakamatsu, Japan, 9/23/14. https://doi.org/10.1109/MCSoC.2014.43
Ohshima S, Katagiri T, Matsumoto M. Performance optimization of SpMV using CRS format by considering OpenMP scheduling on CPUs and MIC. In Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 253-260. 6949479. (Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014). https://doi.org/10.1109/MCSoC.2014.43
Ohshima, Satoshi ; Katagiri, Takahiro ; Matsumoto, Masaharu. / Performance optimization of SpMV using CRS format by considering OpenMP scheduling on CPUs and MIC. Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 253-260 (Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014).
@inproceedings{8df074e3a4fe4897be92a53c0a74dd5f,
title = "Performance optimization of SpMV using CRS format by considering OpenMP scheduling on CPUs and MIC",
abstract = "In this study, we evaluate the performance of sparse matrix-vector multiplication (SpMV) using the compressed row storage (CRS) format on CPUs and MIC. We focus on the relationship between OpenMP scheduling and performance. The performance of SpMV is measured using various OpenMP scheduling settings and the results are analyzed, which show that OpenMP scheduling has a considerable effect on the performance of SpMV. We confirm that some scheduling settings resulted in performance improvements compared with default scheduling for particular matrices. The results of the evaluation show that the performance of SpMV is improved by up to 1.57 times compared with SPARC64 IXfx, 2.47 times compared with Xeon Ivy Bridge-EP, and 2.26 times compared with Knights Corner. Next, we modify the SpMV function of OpenATLib, an auto-tuned numerical library, to consider the scheduling of optimization as an additional SpMV implementation. We measure the performance of the GMRES solver and obtain performance improvements of up to 11.4{\%}. These results will help to improve the performance of various numerical calculation applications.",
author = "Satoshi Ohshima and Takahiro Katagiri and Masaharu Matsumoto",
year = "2014",
month = "11",
day = "6",
doi = "10.1109/MCSoC.2014.43",
language = "English",
series = "Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "253--260",
booktitle = "Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014",
address = "United States",

}

TY - GEN

T1 - Performance optimization of SpMV using CRS format by considering OpenMP scheduling on CPUs and MIC

AU - Ohshima, Satoshi

AU - Katagiri, Takahiro

AU - Matsumoto, Masaharu

PY - 2014/11/6

Y1 - 2014/11/6

N2 - In this study, we evaluate the performance of sparse matrix-vector multiplication (SpMV) using the compressed row storage (CRS) format on CPUs and MIC. We focus on the relationship between OpenMP scheduling and performance. The performance of SpMV is measured using various OpenMP scheduling settings and the results are analyzed, which show that OpenMP scheduling has a considerable effect on the performance of SpMV. We confirm that some scheduling settings resulted in performance improvements compared with default scheduling for particular matrices. The results of the evaluation show that the performance of SpMV is improved by up to 1.57 times compared with SPARC64 IXfx, 2.47 times compared with Xeon Ivy Bridge-EP, and 2.26 times compared with Knights Corner. Next, we modify the SpMV function of OpenATLib, an auto-tuned numerical library, to consider the scheduling of optimization as an additional SpMV implementation. We measure the performance of the GMRES solver and obtain performance improvements of up to 11.4%. These results will help to improve the performance of various numerical calculation applications.

AB - In this study, we evaluate the performance of sparse matrix-vector multiplication (SpMV) using the compressed row storage (CRS) format on CPUs and MIC. We focus on the relationship between OpenMP scheduling and performance. The performance of SpMV is measured using various OpenMP scheduling settings and the results are analyzed, which show that OpenMP scheduling has a considerable effect on the performance of SpMV. We confirm that some scheduling settings resulted in performance improvements compared with default scheduling for particular matrices. The results of the evaluation show that the performance of SpMV is improved by up to 1.57 times compared with SPARC64 IXfx, 2.47 times compared with Xeon Ivy Bridge-EP, and 2.26 times compared with Knights Corner. Next, we modify the SpMV function of OpenATLib, an auto-tuned numerical library, to consider the scheduling of optimization as an additional SpMV implementation. We measure the performance of the GMRES solver and obtain performance improvements of up to 11.4%. These results will help to improve the performance of various numerical calculation applications.

UR - http://www.scopus.com/inward/record.url?scp=84917732692&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84917732692&partnerID=8YFLogxK

U2 - 10.1109/MCSoC.2014.43

DO - 10.1109/MCSoC.2014.43

M3 - Conference contribution

T3 - Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014

SP - 253

EP - 260

BT - Proceedings - 2014 IEEE 8th International Symposium on Embedded Multicore/Manycore SoCs, MCSoC 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -