Increase the concurrency for multi-core systems through collision array based workload assignment

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

Abstract

Simply upgrading to multi-core systems has been proven to provide only minor speedup compared with single core systems. However, because multi-core systems use shared memory, the bottleneck for speedup lies in the memory I/O. The only way to fully utilize multiple cores is to increase parallelism and concurrency. This paper proposes a new collision array based workload assignment to increase data request cancellation. Through a task flow partitioning algorithm, we minimize sequential data access and then dynamically schedule tasks while increasing the data request cancellations. We have shown experimentally that this method can boost general system throughput by 3 times and reduce the execution time by an average of 27.6%. For a 3×3 multi-core system, if all nine cores are utilized, the speedup can reach 5.0 times. If only three cores are utilized to reduce communication and power, the proposed method can obtain 2.3 times speedup.

Original languageEnglish (US)
Title of host publicationProceedings - 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1209-1215
Number of pages7
Volume2
ISBN (Print)9781479931965
DOIs
StatePublished - Nov 5 2014
Event2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014 - Sapporo City, Hokkaido, Japan
Duration: Apr 26 2014Apr 28 2014

Other

Other2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014
CountryJapan
CitySapporo City, Hokkaido
Period4/26/144/28/14

Fingerprint

Data storage equipment
Throughput
Communication

Keywords

  • collision array
  • concurrent
  • multi-core system
  • workload assignment

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Zhou, H., Powers, L. S., & Wang, M. (2014). Increase the concurrency for multi-core systems through collision array based workload assignment. In Proceedings - 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014 (Vol. 2, pp. 1209-1215). [6947862] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/InfoSEEE.2014.6947862

Increase the concurrency for multi-core systems through collision array based workload assignment. / Zhou, He; Powers, Linda S; Wang, Meiling.

Proceedings - 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014. Vol. 2 Institute of Electrical and Electronics Engineers Inc., 2014. p. 1209-1215 6947862.

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

Zhou, H, Powers, LS & Wang, M 2014, Increase the concurrency for multi-core systems through collision array based workload assignment. in Proceedings - 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014. vol. 2, 6947862, Institute of Electrical and Electronics Engineers Inc., pp. 1209-1215, 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014, Sapporo City, Hokkaido, Japan, 4/26/14. https://doi.org/10.1109/InfoSEEE.2014.6947862
Zhou H, Powers LS, Wang M. Increase the concurrency for multi-core systems through collision array based workload assignment. In Proceedings - 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014. Vol. 2. Institute of Electrical and Electronics Engineers Inc. 2014. p. 1209-1215. 6947862 https://doi.org/10.1109/InfoSEEE.2014.6947862
Zhou, He ; Powers, Linda S ; Wang, Meiling. / Increase the concurrency for multi-core systems through collision array based workload assignment. Proceedings - 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014. Vol. 2 Institute of Electrical and Electronics Engineers Inc., 2014. pp. 1209-1215
@inproceedings{3ee65f5a19d34473b7a050bb93a5d1af,
title = "Increase the concurrency for multi-core systems through collision array based workload assignment",
abstract = "Simply upgrading to multi-core systems has been proven to provide only minor speedup compared with single core systems. However, because multi-core systems use shared memory, the bottleneck for speedup lies in the memory I/O. The only way to fully utilize multiple cores is to increase parallelism and concurrency. This paper proposes a new collision array based workload assignment to increase data request cancellation. Through a task flow partitioning algorithm, we minimize sequential data access and then dynamically schedule tasks while increasing the data request cancellations. We have shown experimentally that this method can boost general system throughput by 3 times and reduce the execution time by an average of 27.6{\%}. For a 3×3 multi-core system, if all nine cores are utilized, the speedup can reach 5.0 times. If only three cores are utilized to reduce communication and power, the proposed method can obtain 2.3 times speedup.",
keywords = "collision array, concurrent, multi-core system, workload assignment",
author = "He Zhou and Powers, {Linda S} and Meiling Wang",
year = "2014",
month = "11",
day = "5",
doi = "10.1109/InfoSEEE.2014.6947862",
language = "English (US)",
isbn = "9781479931965",
volume = "2",
pages = "1209--1215",
booktitle = "Proceedings - 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Increase the concurrency for multi-core systems through collision array based workload assignment

AU - Zhou, He

AU - Powers, Linda S

AU - Wang, Meiling

PY - 2014/11/5

Y1 - 2014/11/5

N2 - Simply upgrading to multi-core systems has been proven to provide only minor speedup compared with single core systems. However, because multi-core systems use shared memory, the bottleneck for speedup lies in the memory I/O. The only way to fully utilize multiple cores is to increase parallelism and concurrency. This paper proposes a new collision array based workload assignment to increase data request cancellation. Through a task flow partitioning algorithm, we minimize sequential data access and then dynamically schedule tasks while increasing the data request cancellations. We have shown experimentally that this method can boost general system throughput by 3 times and reduce the execution time by an average of 27.6%. For a 3×3 multi-core system, if all nine cores are utilized, the speedup can reach 5.0 times. If only three cores are utilized to reduce communication and power, the proposed method can obtain 2.3 times speedup.

AB - Simply upgrading to multi-core systems has been proven to provide only minor speedup compared with single core systems. However, because multi-core systems use shared memory, the bottleneck for speedup lies in the memory I/O. The only way to fully utilize multiple cores is to increase parallelism and concurrency. This paper proposes a new collision array based workload assignment to increase data request cancellation. Through a task flow partitioning algorithm, we minimize sequential data access and then dynamically schedule tasks while increasing the data request cancellations. We have shown experimentally that this method can boost general system throughput by 3 times and reduce the execution time by an average of 27.6%. For a 3×3 multi-core system, if all nine cores are utilized, the speedup can reach 5.0 times. If only three cores are utilized to reduce communication and power, the proposed method can obtain 2.3 times speedup.

KW - collision array

KW - concurrent

KW - multi-core system

KW - workload assignment

UR - http://www.scopus.com/inward/record.url?scp=84913530032&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84913530032&partnerID=8YFLogxK

U2 - 10.1109/InfoSEEE.2014.6947862

DO - 10.1109/InfoSEEE.2014.6947862

M3 - Conference contribution

AN - SCOPUS:84913530032

SN - 9781479931965

VL - 2

SP - 1209

EP - 1215

BT - Proceedings - 2014 International Conference on Information Science, Electronics and Electrical Engineering, ISEEE 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -