,
Frank Slomka
Creative Commons Attribution 4.0 International license
Real-time systems are required to produce correct responses which also need to be on time. Two methods to verify the timeliness of the results are the compositional performance analysis and the real-time calculus. Both methods use bounds as inputs from which they derive bounds for the timeliness. Therefore, it is preferable to have better bounds for the inputs. Any of these bounds may contain pessimism. Because both methods use a modular/compositional approach for distributed real-time systems, the pessimism in the bounds accumulate. Therefore, it is desirable to keep the pessimism as low as possible. In this work we investigate distributed real-time systems where tasks have an activation dependency, i. e., the activation of one task depends on the termination of a job of another task. Specifically, we compare the derived bound of how frequent the jobs of a task terminate and subsequently how frequent the activation dependent task activates and releases its jobs. Our contribution is a formal comparison of these bounds that the compositional performance analysis and the real-time calculus derive by means of a mathematical proof. We show that for identical inputs the real-time calculus derives a bound that is not worse than the bound that the compositional performance analysis derives for a single module/component. The mathematical proof also removes any uncertainty that existing empirical comparisons of these bounds may contain.
@InProceedings{pollex_et_al:LIPIcs.ECRTS.2025.16,
author = {Pollex, Victor and Slomka, Frank},
title = {{Formal Comparison of Outgoing Event Streams Between Compositional Performance Analysis and Real-Time Calculus}},
booktitle = {37th Euromicro Conference on Real-Time Systems (ECRTS 2025)},
pages = {16:1--16:25},
series = {Leibniz International Proceedings in Informatics (LIPIcs)},
ISBN = {978-3-95977-377-5},
ISSN = {1868-8969},
year = {2025},
volume = {335},
editor = {Mancuso, Renato},
publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
address = {Dagstuhl, Germany},
URL = {https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ECRTS.2025.16},
URN = {urn:nbn:de:0030-drops-235945},
doi = {10.4230/LIPIcs.ECRTS.2025.16},
annote = {Keywords: compositional performance analysis, real-time calculus, outgoing event streams}
}