Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Execution time budget assignment for mixed criticality systems

Mohamed Amine Khelassi, Yasmina Abdeddaïm

TL;DR

The paper addresses allocating execution-time budgets to low-criticality tasks in mixed-criticality real-time systems by accounting for execution-time variability. It defines dispersion-based variability, notably $VWCET$, and presents a greedy budget-assignment heuristic that uses $TV_i$ to set LO budgets while assigning HI budgets to their maximum. Evaluations on ARM-Cortex A53 simulations and Zynq hardware demonstrate that the variability-aware approach improves the probability of staying within budgets and meeting high-criticality deadlines, with the optimal solution offering better scores at higher computational cost. The method provides a practical, scheduling-agnostic tool for budget assignment that does not rely on full probabilistic distributions, and shows promise for online adaptation in future work.

Abstract

In this paper we propose to quantify execution time variability of programs using statistical dispersion parameters. We show how the execution time variability can be exploited in mixed criticality real-time systems. We propose a heuristic to compute the execution time budget to be allocated to each low criticality real-time task according to its execution time variability. We show using experiments and simulations that the proposed heuristic reduces the probability of exceeding the allocated budget compared to algorithms which do not take into account the execution time variability parameter.

Execution time budget assignment for mixed criticality systems

TL;DR

The paper addresses allocating execution-time budgets to low-criticality tasks in mixed-criticality real-time systems by accounting for execution-time variability. It defines dispersion-based variability, notably , and presents a greedy budget-assignment heuristic that uses to set LO budgets while assigning HI budgets to their maximum. Evaluations on ARM-Cortex A53 simulations and Zynq hardware demonstrate that the variability-aware approach improves the probability of staying within budgets and meeting high-criticality deadlines, with the optimal solution offering better scores at higher computational cost. The method provides a practical, scheduling-agnostic tool for budget assignment that does not rely on full probabilistic distributions, and shows promise for online adaptation in future work.

Abstract

In this paper we propose to quantify execution time variability of programs using statistical dispersion parameters. We show how the execution time variability can be exploited in mixed criticality real-time systems. We propose a heuristic to compute the execution time budget to be allocated to each low criticality real-time task according to its execution time variability. We show using experiments and simulations that the proposed heuristic reduces the probability of exceeding the allocated budget compared to algorithms which do not take into account the execution time variability parameter.
Paper Structure (8 sections, 2 equations, 5 figures, 4 tables, 1 algorithm)

This paper contains 8 sections, 2 equations, 5 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related work
  3. Time variability quantification
  4. Mixed criticality budget assignment problem
  5. Budget assignment heuristic
  6. Simulation-based evaluation
  7. Benchmark-based evaluation
  8. Conclusion

Figures (5)

  • Figure 1: Execution times (cycles) of programs executed in an ARM Cortex 53
  • Figure 2: Computation time of Opt compared to Algorithm 1
  • Figure 3: Algorithms comparison for Scenario 1
  • Figure 4: Algorithms comparison for Scenario 2
  • Figure 5: Algorithms comparison for Scenario 3

Theorems & Definitions (3)

  • Definition 1: Execution time variability
  • Definition 2: Coefficient of variation to the maximum
  • Definition 3: Mixed criticality schedulability