Recently, Internet technology has been applied to home appliances as well as electronic devices such as PCs, laptops and mobile phones, requiring improved networks with high speed and bandwidth to handle a variety of data. In particular, network management techniques to maintain load balance using Software Defined networking (SDN) are cited as one of the most promising paradigms. In this paper, we propose a Deep Learning Mechanism (DLMBN) mechanism (Deep Learning Mechanism on Blockchain) that optimizes the load balance that can occur in the network by deep learning some important information related to the load balance after connecting the information of multiple distributed controllers into the blockchain. The proposed mechanism binds and manages the load of each controller distributed over the network with a blockchain, thus reducing load time while dynamically balancing the load balance. In particular, deep learning technology was used to ensure that each controller classified as a group would not be biased to one side and would maintain a balanced load balance across the entire network. As a result of the experiment, the proposed mechanism improved the load balance retention time by 14.6% on average compared to the mechanism previously studied, and the efficiency of SDNs processed in multiple groups by 17.3% on average. In addition, the overhead of SDNs for each group was lowered by 7.9%.
목차
Abstract1 Index Terms I. INTRODUCTION II. RELATED WORKS A. Software defined network(SDN) B. Previous Research III. MLMBN MECHANISM OPTIMIZED NETWORK LOAD BALANCING A. Network B. Blockchain-based SDN Controller Configuration C. Create controller information IV. PERFORMANCE EVALUATION A. Experimental Environment B. Load Balance Retention Time C. Efficiency D. Overhead V. CONCLUSION REFERENCES