Open AccessResearch on network load balancing method based on simulated annealing algorithm and genetic algorithm
Author(s) -
Yinghui Xu,
Geng Qing Yang
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1237/2/022137
Subject(s) - simulated annealing , computer science , algorithm , mathematical optimization , network congestion , genetic algorithm , adaptive simulated annealing , burstiness , load balancing (electrical power) , mathematics , computer network , geometry , machine learning , network packet , grid
Due to the high burstiness and high real-time performance of modern network load, it is difficult to meet the requirements of purely simulated annealing algorithm or genetic algorithm for network resource utilization and flow control problems, resulting in low network resource utilization. Congestion is serious. In order to make network load balancing more reasonable, a method of combining simulated annealing algorithm with genetic algorithm is proposed. Firstly, the genetic algorithm is used to globally search the network load balancing problem, so that the solution of the problem is quickly in the vicinity of the global optimal region. Then the simulated annealing algorithm is used to further local optimization near the global optimal region, and the optimal solution for the network load balancing problem is found. The simulation results show that the method improves the global search speed, improves the network resource utilization, and significantly improves the network load imbalance.