You are required to read and agree to the below before accessing a full-text version of an article in the IDE article repository.

The full-text document you are about to access is subject to national and international copyright laws. In most cases (but not necessarily all) the consequence is that personal use is allowed given that the copyright owner is duly acknowledged and respected. All other use (typically) require an explicit permission (often in writing) by the copyright owner.

For the reports in this repository we specifically note that

  • the use of articles under IEEE copyright is governed by the IEEE copyright policy (available at http://www.ieee.org/web/publications/rights/copyrightpolicy.html)
  • the use of articles under ACM copyright is governed by the ACM copyright policy (available at http://www.acm.org/pubs/copyright_policy/)
  • technical reports and other articles issued by M‰lardalen University is free for personal use. For other use, the explicit consent of the authors is required
  • in other cases, please contact the copyright owner for detailed information

By accepting I agree to acknowledge and respect the rights of the copyright owner of the document I am about to access.

If you are in doubt, feel free to contact webmaster@ide.mdh.se

Analytical Approximations in Probabilistic Analysis of Real-Time Systems

Fulltext:


Publication Type:

Conference/Workshop Paper

Venue:

43rd IEEE Real-Time Systems Symposium 2022

DOI:

10.1109/RTSS55097.2022.00023


Abstract

Probabilistic timing and schedulability analysis of real-time systems is constrained by the problem of often intractable exact computations. The intractability problem is present whenever there is a large number of entities to be analysed, e.g., jobs, tasks, etc. In the last few years, the analytical approximations for deadline-miss probability emerged as an important solution in the above problem domain.In this paper, we explore analytical solutions for two major problems that are present in the probabilistic analysis of real-time systems. First, for a safe approximation of the entire probability distributions (e.g., of the accumulated execution workloads), we show how the Berry-Esseen theorem can be used. Second, we propose an approximation built on the Berry-Esseen theorem for efficient computation of the quantile functions of probability execution distributions. We also show the asymptotic bounds on the execution distribution of the fixed-priority preemptive tasks.In the evaluation, we investigate the complexity and accuracy of the proposed methods as the number of analysed jobs and tasks increases. The methods are compared with the circular convolution approach. We also investigate the memory footprint comparison between the proposed Berry-Esseen-based solutions and the circular convolution. The contributions and results presented in this paper complement the state-of-the-art in accurate and efficient probabilistic analysis of real-time systems.

Bibtex

@inproceedings{Markovic6546,
author = {Filip Markovic and Thomas Nolte and Alessandro Papadopoulos},
title = {Analytical Approximations in Probabilistic Analysis of Real-Time Systems},
pages = {158--171},
month = {December},
year = {2022},
booktitle = {43rd IEEE Real-Time Systems Symposium 2022},
url = {http://www.es.mdu.se/publications/6546-}
}