Two Parameter Workload Characterization for Improved Dataflow Analysis Accuracy

J.P.H.M. Hausmans, S.J. Geuns, M.H. Wiggers, Marco Jan Gerrit Bekooij

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

7 Citations (Scopus)
566 Downloads (Pure)

Abstract

Real-time stream processing applications, such as radios, can often be modeled intuitively with dataflow models. Given the Worst-Case Execution Times (WCETs) of the tasks, which characterizes workload with one parameter, dataflow analysis techniques have been used to compute the minimum throughput and maximum latency of these applications. However, a large difference between the WCETs of the tasks and their average execution times can result in a large difference between the computed worst-case throughput and the actual obtained throughput. To reduce the difference between the worst-case throughput, determined by analysis, and the actual obtained throughput, we introduce in this paper a two parameter (σ,Ͽ) workload characterization of the tasks to improve the accuracy of dataflow analysis. The (σ, Ͽ) workload characterization captures information on the maximum cumulative execution time of consecutive executions of a task and can therefore be seen as a generalization of the WCET characterization. We show how the (σ,Ͽ) workload characterization can be used in combination with several types of dataflow graphs and how it can be used to improve the temporal analysis results of real-time stream processing applications. We illustrate this for a DVB-T radio application, a car-radio application and a data-dependent MP3 playback application.
Original languageUndefined
Title of host publicationProceedings of the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS)
Place of PublicationWashington, DC, USA
PublisherIEEE Computer Society
Pages117-126
Number of pages10
ISBN (Print)978-1-4799-0186-9
DOIs
Publication statusPublished - 10 Apr 2013

Publication series

Name
PublisherIEEE Computer Society
ISSN (Print)1080-1812

Keywords

  • METIS-297742
  • EWI-23534
  • IR-86747

Cite this

Hausmans, J. P. H. M., Geuns, S. J., Wiggers, M. H., & Bekooij, M. J. G. (2013). Two Parameter Workload Characterization for Improved Dataflow Analysis Accuracy. In Proceedings of the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS) (pp. 117-126). Washington, DC, USA: IEEE Computer Society. https://doi.org/10.1109/RTAS.2013.6531085
Hausmans, J.P.H.M. ; Geuns, S.J. ; Wiggers, M.H. ; Bekooij, Marco Jan Gerrit. / Two Parameter Workload Characterization for Improved Dataflow Analysis Accuracy. Proceedings of the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS). Washington, DC, USA : IEEE Computer Society, 2013. pp. 117-126
@inproceedings{19ddcb2e98d5459b85d890d44a66d796,
title = "Two Parameter Workload Characterization for Improved Dataflow Analysis Accuracy",
abstract = "Real-time stream processing applications, such as radios, can often be modeled intuitively with dataflow models. Given the Worst-Case Execution Times (WCETs) of the tasks, which characterizes workload with one parameter, dataflow analysis techniques have been used to compute the minimum throughput and maximum latency of these applications. However, a large difference between the WCETs of the tasks and their average execution times can result in a large difference between the computed worst-case throughput and the actual obtained throughput. To reduce the difference between the worst-case throughput, determined by analysis, and the actual obtained throughput, we introduce in this paper a two parameter (σ,Ͽ) workload characterization of the tasks to improve the accuracy of dataflow analysis. The (σ, Ͽ) workload characterization captures information on the maximum cumulative execution time of consecutive executions of a task and can therefore be seen as a generalization of the WCET characterization. We show how the (σ,Ͽ) workload characterization can be used in combination with several types of dataflow graphs and how it can be used to improve the temporal analysis results of real-time stream processing applications. We illustrate this for a DVB-T radio application, a car-radio application and a data-dependent MP3 playback application.",
keywords = "METIS-297742, EWI-23534, IR-86747",
author = "J.P.H.M. Hausmans and S.J. Geuns and M.H. Wiggers and Bekooij, {Marco Jan Gerrit}",
note = "10.1109/RTAS.2013.6531085",
year = "2013",
month = "4",
day = "10",
doi = "10.1109/RTAS.2013.6531085",
language = "Undefined",
isbn = "978-1-4799-0186-9",
publisher = "IEEE Computer Society",
pages = "117--126",
booktitle = "Proceedings of the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS)",
address = "United States",

}

Hausmans, JPHM, Geuns, SJ, Wiggers, MH & Bekooij, MJG 2013, Two Parameter Workload Characterization for Improved Dataflow Analysis Accuracy. in Proceedings of the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS). IEEE Computer Society, Washington, DC, USA, pp. 117-126. https://doi.org/10.1109/RTAS.2013.6531085

Two Parameter Workload Characterization for Improved Dataflow Analysis Accuracy. / Hausmans, J.P.H.M.; Geuns, S.J.; Wiggers, M.H.; Bekooij, Marco Jan Gerrit.

Proceedings of the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS). Washington, DC, USA : IEEE Computer Society, 2013. p. 117-126.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Two Parameter Workload Characterization for Improved Dataflow Analysis Accuracy

AU - Hausmans, J.P.H.M.

AU - Geuns, S.J.

AU - Wiggers, M.H.

AU - Bekooij, Marco Jan Gerrit

N1 - 10.1109/RTAS.2013.6531085

PY - 2013/4/10

Y1 - 2013/4/10

N2 - Real-time stream processing applications, such as radios, can often be modeled intuitively with dataflow models. Given the Worst-Case Execution Times (WCETs) of the tasks, which characterizes workload with one parameter, dataflow analysis techniques have been used to compute the minimum throughput and maximum latency of these applications. However, a large difference between the WCETs of the tasks and their average execution times can result in a large difference between the computed worst-case throughput and the actual obtained throughput. To reduce the difference between the worst-case throughput, determined by analysis, and the actual obtained throughput, we introduce in this paper a two parameter (σ,Ͽ) workload characterization of the tasks to improve the accuracy of dataflow analysis. The (σ, Ͽ) workload characterization captures information on the maximum cumulative execution time of consecutive executions of a task and can therefore be seen as a generalization of the WCET characterization. We show how the (σ,Ͽ) workload characterization can be used in combination with several types of dataflow graphs and how it can be used to improve the temporal analysis results of real-time stream processing applications. We illustrate this for a DVB-T radio application, a car-radio application and a data-dependent MP3 playback application.

AB - Real-time stream processing applications, such as radios, can often be modeled intuitively with dataflow models. Given the Worst-Case Execution Times (WCETs) of the tasks, which characterizes workload with one parameter, dataflow analysis techniques have been used to compute the minimum throughput and maximum latency of these applications. However, a large difference between the WCETs of the tasks and their average execution times can result in a large difference between the computed worst-case throughput and the actual obtained throughput. To reduce the difference between the worst-case throughput, determined by analysis, and the actual obtained throughput, we introduce in this paper a two parameter (σ,Ͽ) workload characterization of the tasks to improve the accuracy of dataflow analysis. The (σ, Ͽ) workload characterization captures information on the maximum cumulative execution time of consecutive executions of a task and can therefore be seen as a generalization of the WCET characterization. We show how the (σ,Ͽ) workload characterization can be used in combination with several types of dataflow graphs and how it can be used to improve the temporal analysis results of real-time stream processing applications. We illustrate this for a DVB-T radio application, a car-radio application and a data-dependent MP3 playback application.

KW - METIS-297742

KW - EWI-23534

KW - IR-86747

U2 - 10.1109/RTAS.2013.6531085

DO - 10.1109/RTAS.2013.6531085

M3 - Conference contribution

SN - 978-1-4799-0186-9

SP - 117

EP - 126

BT - Proceedings of the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS)

PB - IEEE Computer Society

CY - Washington, DC, USA

ER -

Hausmans JPHM, Geuns SJ, Wiggers MH, Bekooij MJG. Two Parameter Workload Characterization for Improved Dataflow Analysis Accuracy. In Proceedings of the 2013 IEEE 19th Real-Time and Embedded Technology and Applications Symposium (RTAS). Washington, DC, USA: IEEE Computer Society. 2013. p. 117-126 https://doi.org/10.1109/RTAS.2013.6531085