Work

Enhanced Particle Swarm Optimization for Automated Compatibility Testing in Cloud-based Distributed Systems

Year: 2025

Type: article

Abstract: Ensuring software compatibility in cloud-based distributed systems presents significant challenges due to theheterogeneous nature of cloud environments and the complexity of distributed architectures. This paper proposes anenhanced particle swarm optimization (PSO) approach for automated compatibility testing that addresses the limitationsof traditional testing methods. The methodology integrates improved PSO algorithms with TLA+ formal verificationand Jepsen distributed testing frameworks, incorporating dispersion adjustment mechanisms to prevent prematureconvergence and enhance testing coverage. Key improvements include adaptive weight adjustment, collision radiusoptimization, and fitness function modification based on branch path coverage metrics. Experimental validationdemonstrates significant improvements in compatibility testing efficiency, coverage breadth, and system robustnessacross diverse cloud computing environments. The proposed approach effectively optimizes software performance andreliability while ensuring seamless operation of distributed systems in dynamic cloud infrastructures.

Source: Journal of Cross-Disciplinary Sustainability Innovations

DOI: 10.64229/c54gew59

Authors: Dai Li, Weishuo Lan

Institutions: Rutgers, The State University of New Jersey, University of Rochester

Language: English

Topic: Software Testing and Debugging Techniques

Subfield: Software

Field: Computer Science

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: diamond

OpenAlex ID: W4412847241

FWCI: 2.45

Cites:

Cited by: 1

Related to: 10