An Enhanced Cloud Allocation Approach based on Metaheuristics Algorithms

Document Type : Original full papers (regular papers)

Authors

1 Information Systems Department.

2 Information Systems Department, Faculty of Computers and Artificial Intelligence Helwan University Helwan, Egypt

3 Faculty of Computers and Artificial Intelligence, Fayoum University, Fayoum Egypt.

Abstract

In cloud computing environments, effective data placement is critical for optimizing system performance and resource utilization. This research introduces an innovative framework: Enhanced Cloud Data Placement Strategy Using Marine Predator Optimization. This algorithm leverages the Marine Predators Algorithm (MPA) -a nature-inspired metaheuristic inspired by marine predators' hunting behavior, balancing exploration and exploitation for efficient optimization-, to address theis challenge. The proposed framework leverages MPA’s exploration and exploitation capabilities to reduce data movement between data centers, and enhance resource allocation efficiency. Through simulation in a controlled environment using the CloudSim toolkit, we evaluated the performance of MPA in comparison with other state-of-the-art metaheuristic algorithms, including the Gaining Sharing Knowledge-based Algorithm (GSKA), War Strategy Optimization (WarSO), Generalized Whale Optimization Algorithm (GWO_WOA), and Success History Intelligent Optimizer (SHIO). Experimental results demonstrate that MPA outperforms these algorithms in terms of runtime, and overall resource utilization. Further tests, including scalability evaluations with increasing dataset sizes and data center numbers, revealed MPA’s robustness and adaptability for large-scale cloud infrastructures. The performance comparison indicates that applying MPA to solve the proposed problems consistently yields lower makespan and runtime, positioning it as a promising solution for dynamic and heterogeneous cloud environments.

Keywords

Main Subjects

Files

Share

How to cite

Statistics