DROPS

Document

DOI: 10.4230/OASIcs.WCET.2018.4

Toward Contention Analysis for Parallel Executing Real-Time Tasks

Authors: Fabrice Guet, Luca Santinelli, Jérôme Morio, Guillaume Phavorin, and Eric Jenn

Published in: OASIcs, Volume 63, 18th International Workshop on Worst-Case Execution Time Analysis (WCET 2018)

Abstract

In measurement-based probabilistic timing analysis, the execution conditions imposed to tasks as measurement scenarios, have a strong impact to the worst-case execution time estimates. The scenarios and their effects on the task execution behavior have to be deeply investigated. The aim has to be to identify and to guarantee the scenarios that lead to the maximum measurements, i.e. the worst-case scenarios, and use them to assure the worst-case execution time estimates. We propose a contention analysis in order to identify the worst contentions that a task can suffer from concurrent executions. The work focuses on the interferences on shared resources (cache memories and memory buses) from parallel executions in multi-core real-time systems. Our approach consists of searching for possible task contenders for parallel executions, modeling their contentiousness, and classifying the measurement scenarios accordingly. We identify the most contentious ones and their worst-case effects on task execution times. The measurement-based probabilistic timing analysis is then used to verify the analysis proposed, qualify the scenarios with contentiousness, and compare them. A parallel execution simulator for multi-core real-time system is developed and used for validating our framework. The framework applies heuristics and assumptions that simplify the system behavior. It represents a first step for developing a complete approach which would be able to guarantee the worst-case behavior.

Cite as

Fabrice Guet, Luca Santinelli, Jérôme Morio, Guillaume Phavorin, and Eric Jenn. Toward Contention Analysis for Parallel Executing Real-Time Tasks. In 18th International Workshop on Worst-Case Execution Time Analysis (WCET 2018). Open Access Series in Informatics (OASIcs), Volume 63, pp. 4:1-4:13, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2018)

Copy BibTex To Clipboard

@InProceedings{guet_et_al:OASIcs.WCET.2018.4,
  author =	{Guet, Fabrice and Santinelli, Luca and Morio, J\'{e}r\^{o}me and Phavorin, Guillaume and Jenn, Eric},
  title =	{{Toward Contention Analysis for Parallel Executing Real-Time Tasks}},
  booktitle =	{18th International Workshop on Worst-Case Execution Time Analysis (WCET 2018)},
  pages =	{4:1--4:13},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-073-6},
  ISSN =	{2190-6807},
  year =	{2018},
  volume =	{63},
  editor =	{Brandner, Florian},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.WCET.2018.4},
  URN =		{urn:nbn:de:0030-drops-97506},
  doi =		{10.4230/OASIcs.WCET.2018.4},
  annote =	{Keywords: Contention analysis, parallel executions, measurement-based probabilistic timing analysis, probabilistic worst-case execution time}
}

Document

DOI: 10.4230/OASIcs.WCET.2017.3

On the Representativity of Execution Time Measurements: Studying Dependence and Multi-Mode Tasks

Authors: Fabrice Guet, Luca Santinelli, and Jerome Morio

Published in: OASIcs, Volume 57, 17th International Workshop on Worst-Case Execution Time Analysis (WCET 2017)

Abstract

The Measurement-Based Probabilistic Timing Analysis (MBPTA) infers probabilistic Worst-Case Execution Time (pWCET) estimates from measurements of tasks execution times; the Extreme Value Theory (EVT) is the statistical tool that MBPTA applies for inferring worst-cases from observations/measurements of the actual task behavior. MBPTA and EVT capability of estimating safe/pessimistic pWCET rely on the quality of the measurements; in particular, execution time measurements have to be representative of the actual system execution conditions and have to cover multiple possible execution conditions. In this work, we investigate statistical dependences between execution time measurements and tasks with multiple runtime operational modes. In the first case, we outline the effects of dependences on the EVT applicability as well as on the quality of the pWCET estimates. In the second case, we propose the best approaches to account for the different task execution modes and guaranteeing safe pWCET estimates that cover them all. The solutions proposed are validated with test cases.

Cite as

Fabrice Guet, Luca Santinelli, and Jerome Morio. On the Representativity of Execution Time Measurements: Studying Dependence and Multi-Mode Tasks. In 17th International Workshop on Worst-Case Execution Time Analysis (WCET 2017). Open Access Series in Informatics (OASIcs), Volume 57, pp. 3:1-3:13, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2017)

Copy BibTex To Clipboard

@InProceedings{guet_et_al:OASIcs.WCET.2017.3,
  author =	{Guet, Fabrice and Santinelli, Luca and Morio, Jerome},
  title =	{{On the Representativity of Execution Time Measurements: Studying Dependence and Multi-Mode Tasks}},
  booktitle =	{17th International Workshop on Worst-Case Execution Time Analysis (WCET 2017)},
  pages =	{3:1--3:13},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-057-6},
  ISSN =	{2190-6807},
  year =	{2017},
  volume =	{57},
  editor =	{Reineke, Jan},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/OASIcs.WCET.2017.3},
  URN =		{urn:nbn:de:0030-drops-73084},
  doi =		{10.4230/OASIcs.WCET.2017.3},
  annote =	{Keywords: Measurement-Based Probabilistic Timing Analysis, probabilistic Worst-Case Execution Time, Extreme Value Theory, Execution Time Measurements Representa}
}

Search Results

Documents authored by Guet, Fabrice

Toward Contention Analysis for Parallel Executing Real-Time Tasks

Abstract

Cite as

On the Representativity of Execution Time Measurements: Studying Dependence and Multi-Mode Tasks

Abstract

Cite as

Thanks for your feedback!

Could not send message