System observation of blocking, non-blocking, and cascading events for runtime monitoring of real-time systems

Jong Chul Lee, Roman L Lysecky

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

3 Citations (Scopus)

Abstract

The complexity of multitasked applications in real-time embedded systems presents key challenges in reliability of task execution. Interactions between periodic and aperiodic tasks can incur unpredictable deviations from ideal execution times. Runtime observations can provide visibility for analyzing real-time execution behavior of vulnerable tasks and detect when such deviations may lead to system failure - potentially allowing correction or failsafe mechanism to be utilized. However, such runtime observation must be non-intrusive as even small perturbations to the system execution can significantly impact the system. This paper presents a system observation framework for monitoring and analyzing rapidly occurring software events at runtime. The system observation framework provides runtime support for defining and controlling software events with using blocking, non-blocking, and cascading event probes.

Original languageEnglish (US)
Title of host publicationProceedings of the International Symposium and Workshop on Engineering of Computer Based Systems
Pages49-58
Number of pages10
DOIs
StatePublished - 2013
Event20th Annual IEEE International Conference and Workshops on the Engineering of Computer Based Systems, ECBS 2013 - Phoenix, AZ, United States
Duration: Apr 22 2013Apr 24 2013

Other

Other20th Annual IEEE International Conference and Workshops on the Engineering of Computer Based Systems, ECBS 2013
CountryUnited States
CityPhoenix, AZ
Period4/22/134/24/13

Fingerprint

Real time systems
Embedded systems
Visibility
Monitoring

Keywords

  • observability
  • online system monitoring
  • testing
  • validation
  • verification

ASJC Scopus subject areas

  • Computer Science(all)
  • Control and Systems Engineering

Cite this

Lee, J. C., & Lysecky, R. L. (2013). System observation of blocking, non-blocking, and cascading events for runtime monitoring of real-time systems. In Proceedings of the International Symposium and Workshop on Engineering of Computer Based Systems (pp. 49-58). [6601572] https://doi.org/10.1109/ECBS.2013.8

System observation of blocking, non-blocking, and cascading events for runtime monitoring of real-time systems. / Lee, Jong Chul; Lysecky, Roman L.

Proceedings of the International Symposium and Workshop on Engineering of Computer Based Systems. 2013. p. 49-58 6601572.

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

Lee, JC & Lysecky, RL 2013, System observation of blocking, non-blocking, and cascading events for runtime monitoring of real-time systems. in Proceedings of the International Symposium and Workshop on Engineering of Computer Based Systems., 6601572, pp. 49-58, 20th Annual IEEE International Conference and Workshops on the Engineering of Computer Based Systems, ECBS 2013, Phoenix, AZ, United States, 4/22/13. https://doi.org/10.1109/ECBS.2013.8
Lee JC, Lysecky RL. System observation of blocking, non-blocking, and cascading events for runtime monitoring of real-time systems. In Proceedings of the International Symposium and Workshop on Engineering of Computer Based Systems. 2013. p. 49-58. 6601572 https://doi.org/10.1109/ECBS.2013.8
Lee, Jong Chul ; Lysecky, Roman L. / System observation of blocking, non-blocking, and cascading events for runtime monitoring of real-time systems. Proceedings of the International Symposium and Workshop on Engineering of Computer Based Systems. 2013. pp. 49-58
@inproceedings{cc89f710a12c47feadea0a32499a8aed,
title = "System observation of blocking, non-blocking, and cascading events for runtime monitoring of real-time systems",
abstract = "The complexity of multitasked applications in real-time embedded systems presents key challenges in reliability of task execution. Interactions between periodic and aperiodic tasks can incur unpredictable deviations from ideal execution times. Runtime observations can provide visibility for analyzing real-time execution behavior of vulnerable tasks and detect when such deviations may lead to system failure - potentially allowing correction or failsafe mechanism to be utilized. However, such runtime observation must be non-intrusive as even small perturbations to the system execution can significantly impact the system. This paper presents a system observation framework for monitoring and analyzing rapidly occurring software events at runtime. The system observation framework provides runtime support for defining and controlling software events with using blocking, non-blocking, and cascading event probes.",
keywords = "observability, online system monitoring, testing, validation, verification",
author = "Lee, {Jong Chul} and Lysecky, {Roman L}",
year = "2013",
doi = "10.1109/ECBS.2013.8",
language = "English (US)",
pages = "49--58",
booktitle = "Proceedings of the International Symposium and Workshop on Engineering of Computer Based Systems",

}

TY - GEN

T1 - System observation of blocking, non-blocking, and cascading events for runtime monitoring of real-time systems

AU - Lee, Jong Chul

AU - Lysecky, Roman L

PY - 2013

Y1 - 2013

N2 - The complexity of multitasked applications in real-time embedded systems presents key challenges in reliability of task execution. Interactions between periodic and aperiodic tasks can incur unpredictable deviations from ideal execution times. Runtime observations can provide visibility for analyzing real-time execution behavior of vulnerable tasks and detect when such deviations may lead to system failure - potentially allowing correction or failsafe mechanism to be utilized. However, such runtime observation must be non-intrusive as even small perturbations to the system execution can significantly impact the system. This paper presents a system observation framework for monitoring and analyzing rapidly occurring software events at runtime. The system observation framework provides runtime support for defining and controlling software events with using blocking, non-blocking, and cascading event probes.

AB - The complexity of multitasked applications in real-time embedded systems presents key challenges in reliability of task execution. Interactions between periodic and aperiodic tasks can incur unpredictable deviations from ideal execution times. Runtime observations can provide visibility for analyzing real-time execution behavior of vulnerable tasks and detect when such deviations may lead to system failure - potentially allowing correction or failsafe mechanism to be utilized. However, such runtime observation must be non-intrusive as even small perturbations to the system execution can significantly impact the system. This paper presents a system observation framework for monitoring and analyzing rapidly occurring software events at runtime. The system observation framework provides runtime support for defining and controlling software events with using blocking, non-blocking, and cascading event probes.

KW - observability

KW - online system monitoring

KW - testing

KW - validation

KW - verification

UR - http://www.scopus.com/inward/record.url?scp=84885216258&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885216258&partnerID=8YFLogxK

U2 - 10.1109/ECBS.2013.8

DO - 10.1109/ECBS.2013.8

M3 - Conference contribution

SP - 49

EP - 58

BT - Proceedings of the International Symposium and Workshop on Engineering of Computer Based Systems

ER -