On the Equivalence of Maximum Reaction Time and Maximum Data Age for Cause-Effect Chains

Real-time systems require a formal guarantee of timing-constraints, not only for individual tasks but also for data-propagation. The timing behavior of data-propagation paths in a given system is typically described by its maximum reaction time and its maximum data age. This paper shows that they are equivalent.
To reach this conclusion, partitioned job chains are introduced, which consist of one immediate forward and one immediate backward job chain. Such partitioned job chains are proven to describe maximum reaction time and maximum data age in a universal manner. This universal description does not only show the equivalence of maximum reaction time and maximum data age, but can also be exploited to speed up the computation of such significantly. In particular, the speed-up for synthesized task sets based on automotive benchmarks can be up to 1600.
Since only very few non-restrictive assumptions are made, the equivalence of maximum data age and maximum reaction time holds for almost any scheduling mechanism and even for tasks which do not adhere to the typical periodic or sporadic task model. This observation is supported by a simulation of a ROS2 navigation system.

  • Published in:
    35th Euromicro Conference on Real-Time Systems (ECRTS 2023)
  • Type:
    Inproceedings
  • Authors:
    Günzel, Mario; Teper, Harun; Chen, Kuan-Hsun; von der Brüggen, Georg; Chen, Jian-Jia
  • Year:
    2023
  • Source:
    https://drops.dagstuhl.de/opus/volltexte/2023/18039

Citation information

Günzel, Mario; Teper, Harun; Chen, Kuan-Hsun; von der Brüggen, Georg; Chen, Jian-Jia: On the Equivalence of Maximum Reaction Time and Maximum Data Age for Cause-Effect Chains, 35th Euromicro Conference on Real-Time Systems (ECRTS 2023), 2023, https://drops.dagstuhl.de/opus/volltexte/2023/18039, Guenzel.etal.2023b,

Associated Lamarr Researchers

lamarr institute person Chen Jian Jia - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Prof. Dr. Jian-Jia Chen

Area Chair Resource-aware ML to the profile