Energy-efficient task offloading and efficient resource allocation for edge computing: a quantum inspired particle swarm optimization approach

Banavath Balaji Naik

doi:10.1007/s10586-024-04833-5

2025-01-21

PRE

Banavath Balaji Naik *

Bollu Priyanka

Md. Sarfaraj Alam Ansari

DOI：10.1007/s10586-024-04833-5

原文链接

原文求助

摘要

En 中文

Efficient processing of workflow applications (WAs) is crucial in edge computing environments to enhance efficiency, flexibility, collaboration, and cost savings. Workflow scheduling involves finding an optimal schedule for a group of sub-tasks while maintaining dependency constraints. It is also a non-deterministic polynomial (NP)-complete problem. One of the most challenging aspects of scheduling workflow applications is generating a valid sequence of execution while adhering to dependency constraints among all sub-tasks. In this paper, we propose a novel energy-effective workflow scheduling algorithm based on Quantum-Inspired Particle Swarm Optimization (QIPSO), termed QIPSO-WSA. This algorithm addresses the challenges of efficiently scheduling workflows while optimizing energy consumption. It leverages concepts from QIPSO to enhance the scheduling process and achieve energy efficiency in edge computing environments. QIPSO-WSA considers several critical factors, including makespan, energy consumption, and resource utilization. Quantum particles (QPs) are generated using quantum bits and are updated using a quantum angle. The QPs are decoded using novel hashing techniques, and the fitness function is designed by incorporating various objectives. Extensive simulations are performed and compared with evolutionary techniques. Statistical analyses, including analysis of variance and the Friedman test, are conducted, and the Taguchi parametric statistical technique is applied to assess performance. The simulation results demonstrate that QIPSO-WSA outperforms existing approaches, achieving improvements in makespan by 11.11%, resource utilization by 5.79%, and energy consumption by 9.52%.

Keyword:

Edge computing

Quantum inspired particle swarm optimization

Quantum particles

Hashing

Workflow scheduling

Energy consumption