Research On Adaptive Resource Scheduling Technology For Business Awareness In Software Defined Mesh Network

Because the existing network can no longer meet the increasing business needs of people,many researchers have proposed to combine Software Defined Network(SDN)with the existing network architecture in order to expand the network scale.Thus,with the widespread application and deployment of Wireless Mesh Networks(WMNs)in daily life,the application of SDN to WMNs to form Software Defined Wireless Mesh Networks has become a new direction of research.However,the deployment of multiple controllers in the SDWMN can easily cause a controller to be overloaded due to receiving a large number of pending data packets at a certain time,which will cause the control layer to be unbalanced.This issue requires attention.At the same time,at the data level,when there are a large number of data streams in the network,how to realize reasonable data flow transmission,avoid network congestion,balance network load,guarantee Qo S and improve resource utilization is also problem to be solved.This article will work in accordance with the above questions:1.Aiming at the problem of imbalance in the control plane caused by the heavy load of some controllers in the network,an SDWMN controller load balancing mapping model is proposed.The model is composed of six modules: topology management module,service awareness module,migration strategy module,and so on.And this article proposes a Switch Migration Load Balancing(SMLB)strategy that comprehensively considers the load balancing degree,migration cost,and number of switches to be migrated.The objective is to complete the migration by changing the master controller of the eligible switch under the overload controller,and reducing the load of the controller to balance the entire control plane.At the same time,an improved Max-Min Ant Colony(IMMAC)algorithm is introduced to optimize the SMLB strategy to obtain a better solution and achieve network equilibrium.2.In view of the fact that the transmission of multiple network data streams is likely to cause local link congestion at the data level in the SDWMN,a Code-aware Load-Balancing routing model with guaranteed Qo S(QCLB)is designed.The model includes: initial path finding module,network information collection module,service awareness module,routing measurement module,routing management module.The QCLB routing strategy comprehensively considers the impact of packet loss rate,coding opportunities,load and interference factors on network links to select the best forwarding path;then this article uses the reverse auction game theory to optimize the QCLB routing algorithm,which can find a better forwarding path,reduce the packet loss rate of the link,improve the utilization of network resources,and achieve load balancing at the network forwarding layer.