Hummingbird-Inspired Modified Particle Swarm Optimization for Efficient Task Scheduling in Cloud Computing

Cloud computing delivers on-demand services and scalable computing power in near real-time, redefining modern computing paradigms. Effective task scheduling remains a critical challenge due to dynamic and heterogeneous workloads, directly influencing energy efficiency, response time, and resource ut...

Full description

Saved in:
Bibliographic Details
Main Authors: Longyang Du, Qingxuan Wang
Format: Article
Language:English
Published: Tamkang University Press 2025-05-01
Series:Journal of Applied Science and Engineering
Subjects:
Online Access:http://jase.tku.edu.tw/articles/jase-202512-28-12-0006
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Cloud computing delivers on-demand services and scalable computing power in near real-time, redefining modern computing paradigms. Effective task scheduling remains a critical challenge due to dynamic and heterogeneous workloads, directly influencing energy efficiency, response time, and resource utilization. The present research presents an enhanced Particle Swarm Optimization (PSO) algorithm inspired by specific hummingbird flight characteristics, chosen for their exceptional agility and efficiency. Five hummingbirdinspired concepts are integrated into PSO: incremental position updates to enhance convergence accuracy, stepwise position changes to avoid local optima, energy-conserving movements reducing computational overhead, decentralized exploration to maintain diversity, and multidirectional searches enhancing solution coverage. Comparative experiments conducted on synthetic and real-world datasets (HPC2N) with diverse task loads demonstrate measurable performance improvements, including up to 18% better resource utilization, up to a 35% decrease in imbalance degree, and up to a 20% improvement in execution cost compared to recent algorithms. These results confirm that each hummingbird-inspired concept distinctly contributes to overcoming conventional PSO limitations, significantly enhancing exploration ability, convergence speed, load balancing, and adaptability to diverse cloud computing scenarios.
ISSN:2708-9967
2708-9975