Research Overview

Task scheduling in cloud computing is a multi-objective problem. A practical scheduler must simultaneously control completion time, resource cost, energy consumption, load balance, and security. Optimizing only one objective can degrade the others. For example, minimizing execution time alone may increase infrastructure cost or energy use.

Proposed Framework

This research presents a hybrid framework that integrates the capabilities of Genetic Algorithm, Particle Swarm Optimization, Firefly Algorithm, and Ant Colony Optimization within a multi-objective scheduling process. The combination is intended to benefit from global exploration, rapid convergence, population diversity, and reinforcement of promising paths in the solution space.

Blockchain Layer

To improve transparency and auditability, task allocation decisions and resource billing records are registered through smart contracts on a private Ethereum network. Once recorded, the allocation history cannot be silently modified, allowing involved parties to examine how resources were assigned and how costs were calculated. Blockchain does not replace the optimization algorithm; it adds a trust and immutable-record layer to the scheduler's output.

Evaluation Environment

The experimental evaluation was conducted using CloudSim, and the proposed framework was compared with established scheduling methods. The reported evaluation criteria include makespan, cost, energy consumption, security score, convergence speed, solution quality, and blockchain communication overhead.

Reported Results

According to the article, the proposed framework achieved a 23% reduction in makespan, 18% cost savings, and a 15% reduction in energy consumption compared with traditional methods. A 99.2% security score was reported, while the communication overhead introduced by blockchain integration was 2.8%. Under high-load scenarios, the hybrid method also showed a 35% improvement in convergence speed and a 28% improvement in solution quality compared with single-objective approaches.

Practical Significance

The main contribution lies in connecting three areas: multi-objective optimization, evolutionary and swarm-based algorithms, and auditable decision recording through smart contracts. This architecture is relevant to cloud providers, cloud-based financial systems, shared academic infrastructure, and environments where billing transparency and trust between multiple parties are important.

Publication Note

This page provides an introduction and analytical summary of the article. For the publisher version, bibliographic details, and full-text access status, use the DOI or ResearchGate links provided on this page.