A Hybrid Meta-Heuristic Algorithm of Load Balancing for Cloud-based Railway Interlocking System

Cloud-based railway interlocking system is a novel solution for interlocking systems with the development of cloud computing. The load balancing problem caused by imbalanced resources in computing platforms is an essential problem in realizing cloud-based railway interlocking system. Conventional load balancing algorithms focus on makespan, throughput and costs. However, a load balancing algorithm for a cloud-based railway interlocking system should focus on optimizing resource utilization and execution time. To this end, we propose a hybrid meta-heurisite algorithm called Improved Genetic Annealing Algorithm (IGAA). The proposed algorithm can guarantee global optimization ability by using the Metropolis criterion and can converge rapidly due to the parallel searching characteristic of the Genetic Algorithm. Finally, the feasibility and effectiveness of the proposed algorithm are verified by experiments.