Efficient task scheduling in cloud computing environment using opposition-based firebug tunicate optimization

doi:10.1007/s10586-026-06561-4

Efficient task scheduling in cloud computing environment using opposition-based firebug tunicate optimization

2026-10-03

0

PRE

AI

Vijayakumar Kedalu Poornachary

K

Kong Fah Tee *

J

Janakiraman Nithiyanantham

DOI：10.1007/s10586-026-06561-4

Original

Original request for help

Save

Abstract

En 中文

Task scheduling in cloud computing is a critical component that directly affects system efficiency, particularly in terms of makespan, resource utilization, and operational cost. However, existing metaheuristic algorithms often suffer from poor balance between exploration and exploitation, leading to premature convergence and suboptimal solutions. These challenges arise due to the dynamic nature of cloud environments, task heterogeneity, and varying Quality of Service (QoS) requirements. To overcome these limitations, this paper proposes a new algorithm named Opposition-Based Firebug Tunicate Optimization (OFTO). The methodology combines the strong local search ability of Firebug Swarm Optimization (FSO) with the robust global search capability of the Tunicate Swarm Optimization (TSO) algorithm. This hybrid approach is further enhanced using Opposition-Based Learning (OBL), which increases population diversity by evaluating both current and opposite candidate solutions during each iteration. This integration ensures comprehensive exploration of the search space while maintaining exploitation efficiency, helping to avoid local optima and accelerate convergence. Furthermore, the proposed algorithm is specifically designed to optimize key Quality of Service (QoS) parameters by minimizing execution cost, reducing makespan, and accelerating convergence making the proposed algorithm a Deadline Sensitive Task Scheduler (DSTS). This deadline-sensitive scheduling capability enables it to assign tasks to the most appropriate virtual machines within a constrained time frame, while ensuring minimal consumption of computational resources. Extensive experimental simulations were conducted, comparing OFTO against several state-of-the-art algorithms including Fault Tolerant Trust-based Harris Hawks Deep Reinforcement Learning Algorithm (FTTHDRLA), Modified-Transfer-Function-Based Binary Particle Swarm Optimization (MTF-BPSO), Rider Cuckoo Optimization Algorithm (RCOA), Tunicate Swarm Optimization (TSO), and Firebug Swarm Optimization (FSO). The results demonstrate that OFTO achieved improvements of 7.96%, 14.98%, 18.50%, 24.75%, and 31.71%, respectively, in terms of overall performance. Additional evaluations, including statistical significance tests such as the T-test, Wilcoxon signed-rank test and Friedman test, confirm the robustness and reliability of the proposed algorithm in achieving superior performance across multiple metrics.

Keywords:

Cloud computing

Convergence-rate

Exploration

Exploitation

Firebug tunicate optimization

Swarm intelligence

Efficient task scheduling in cloud computing environment using opposition-based firebug tunicate optimization

Abstract

Journal

Organization

Cited Papers