Energy-Efficient Mixed-Criticality Multicore Systems

Author First name, Last name, Institution

• Fayyaz Ali, University of Swat
• Saud Wasly, King Abdulaziz University
• Amjad Ali, University of Swat
• Shahid Iqbal, University of Swat
• Asad Masood Khattak, Zayed UniversityFollow
• Bashir Hayat, Institute of Management Sciences
• Shah Khalid, National University of Sciences and Technology

Document Type

Article

Source of Publication

IEEE Access

Publication Date

1-1-2026

Abstract

Balancing energy efficiency with stringent timing guarantees in real-time mixed-criticality systems (MCS) is a key challenge, especially in multicore architectures. This paper introduces a novel energy-aware scheduling framework that integrates dynamic voltage and frequency scaling (DVFS) with a Decreasing-Criticality-Decreasing-Utilization (DCDU) allocation approach. The optimal operating frequencies are obtained at each criticality level; high-criticality tasks are assigned to cores at full operating frequency to maintain timing guarantees, while low-criticality tasks are allocated using worst-case execution times scaled to their optimal frequency. A fixed-priority response-time analysis is used for schedulability in low mode, high mode, and during mode changes. The extensive simulations on randomly generated task sets, with target utilizations ranging from 0.35 to 0.95, show better performance of the proposed approach as compared to mixed criticality power management (MCPM), Random Allocation (RA), Partitioned Ekberg (PEKB), and Naive First Fit (NFF). The proposed approach maintains 100% schedulability up to utilization U= 0.85 and retains around 55% schedulability even at the highest utilization U= 0.95, outperforming RA, PEKB, and NFF by more than 30%. The results demonstrate that integrating DVFS with DCDU provides an energy-efficient solution that significantly enhances schedulability and weighted energy gain compared to other approaches.

DOI Link

10.1109/access.2026.3704317

ISSN

2169-3536

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Volume

14

First Page

92485

Last Page

92500

Disciplines

Computer Sciences

Keywords

Energy-aware scheduling, mixed-criticality systems, multi-core systems, real-time systems, task priorities

Scopus ID

105043220054

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Recommended Citation

Ali, Fayyaz; Wasly, Saud; Ali, Amjad; Iqbal, Shahid; Khattak, Asad Masood; Hayat, Bashir; and Khalid, Shah, "Energy-Efficient Mixed-Criticality Multicore Systems" (2026). All Works. 8176.
https://zuscholars.zu.ac.ae/works/8176

Indexed in Scopus

yes

Open Access

yes

Open Access Type

Gold: This publication is openly available in an open access journal/series