Sign in Subscribe
Machine learning

Researchers Develop Quantum-Inspired Framework for Energy-Efficient Cloud Computing

Strategic Insights

— 1 min read

Researchers at Galgotias University have unveiled a novel Quantum-Inspired Hybrid Reinforcement Learning and Multi-Objective Optimization Framework (QHRMOF) designed to optimize task scheduling, dynamic load balancing, and server consolidation while minimizing power consumption and enhancing system performance. The framework integrates quantum-inspired evolutionary algorithms, hybrid deep reinforcement learning, and multi-objective optimization techniques to achieve significant reductions in energy consumption, makespan, and failure rates.

Key Takeaways:

  • The QHRMOF framework reduces energy consumption by up to 22.84% and makespan by up to 18.76%, while enhancing resource utilization by up to 19.52% and load balancing efficiency by up to 25.39%.
  • The framework uses quantum-inspired evolutionary algorithms to explore the solution space and mitigate local optima, while hybrid deep reinforcement learning forecasts workloads and categorizes virtual machines for efficient task migration and load balancing.
  • Multi-objective optimization reconciles conflicting objectives such as minimizing energy consumption and makespan while maximizing resource utilization and system scalability.
  • Simulations performed on the CloudSim platform using real-world datasets from NASA, HPC2N, and Google workloads indicate that QHRMOF surpasses leading methodologies such as Multi-objective Genetic Algorithm (MOGA), Particle Swarm Optimization (PSO), Deep Reinforcement Learning for Load Balancing (DRL-LB), and Ant Colony Optimization (ACO).

Statistics:

  • QHRMOF reduces energy consumption by up to 22.84%.
  • QHRMOF decreases makespan by up to 18.76%.
  • QHRMOF enhances resource utilization by up to 19.52%.
  • QHRMOF improves load balancing efficiency by up to 25.39%.
  • QHRMOF reduces failure rates by up to 12.67%.

Sources:

QHRMOF: A Quantum-Inspired hybrid Multi-Objective framework for Energy-Efficient task scheduling and load balancing in cloud computing. Journal of Cloud Computing: Advances, Systems and Applications, 2025,14(1):1-38.

https://journalofcloudcomputing.springeropen.com

Umesh Kumar Lilhore, Department of Computer Science and Engineering, Galgotias University, 2025.

Read more