The KD-GATS algorithm: A way for Optimizing Data Replication with Kruskal-Dijkstra and Genetic Tabu Strategy

Data replication in distributed storage networks is a crucial concern due to its suitability for enhancing data access time, reliability, and system performance. Despite the critical need for meta-heuristic algorithms to efficiently address data replication as an NP-hard multi-objective optimization problem, a general algorithm has not been developed due to the reasons that: (i) no single solution technique can efficiently solve all instances within a polynomial time frame; and (ii) the problem requires trade-offs among multiple conflicting objectives, i.e., storage, response time, cost efficiency, and network infrastructure constraints. This paper presents the Kruskal-Dijkstra Genetic Tabu Strategy (KD-GATS) as a novel approach to developing a more efficient solution for efficiently solving data replication in distributed storage networks. The paper will demonstrate the applicability of KD-GATS for minimizing the cost, maximizing storage ratio and number replica created in comparison to the Non-dominated Sorting Genetic Algorithm (NSGA-II) and the Strength Pareto Evolutionary Algorithm (SPEA2). The paper further presents the details of our actual tests with a set of complex network limitations that produced encouraging results for cloud storage.

• data replication
• multi-objective optimization
• metaheuristic algorithms
• storage cost reduction
• storage ratio
• replica placement
• cloud storage
• NSGA-II comparison

• 9 — Industry, innovation and infrastructure

• Cloud Computing and Resource Management
• Advanced Data Storage Technologies
• Metaheuristic Optimization Algorithms Research
• data replication
• multi-objective optimization
• metaheuristic algorithms
• storage cost reduction
• storage ratio
• replica placement
• cloud storage
• NSGA-II comparison
• Information Systems
• Computer Networks and Communications
• Artificial Intelligence
• Computer Science