Research On Compution Collaoration For MEC Nodes And Virtual Machine Migration In Ultra-dense Networks

With the development of mobile devices and new applications grow explosively.People's demand for data traffic and computing power in network increases quickly.In order to solve these problems,the fifth generation mobile communication(5G)has become a key research caught the attention of academic and industrial.As two key technologies in 5G,Ultra-Dense Network(UDN)can provide users sufficient transmission bandwidth to ensure data transfer rate and Mobile Edge Computing(MEC)can provide low-latency computing and caching services.These two technologies have great prospects and have been become two important research directions in communications industry.Deploying Mobile Edge Cloud with Ultra-dense Networks,this technology combines high-speed computing with high throughput to ensure the demand of users.But this combination has also many challenges,such as load balancing,mobility management,interference management and so on.Therefore,the problem of multi-MEC server load balancing and resource scheduling under ultra-dense network need to be solved urgently.Based on this,the load balancing and resource scheduling problem of multi-MEC server in Ultra-dense Networks has become an urgent problem.In this paper,with MEC server in UDN as the main consideration,we propose a MEC node strategy based on cooperative feedback.This strategy mainly realizes the resource sharing between the MEC servers,overcomes the unfairness of the MEC server,and makes full use of the computing resources of the overall server to ensure the service of offloading.Firstly,we establish a system model and this model aims at reducing the time consumption of the task.To ensure the completion time of the task,the model assigns the appropriate server and computing resources for the task by the scheduling algorithm.Then,according to the transmission time and the time consumption design,we design the task collaboration algorithm.The algorithm can calculate the task allocation strategy simply and effectively.Finally,through MATLAB,we carry out the simulation results.The results show that the algorithm is effective in reducing the average time consumption and improving the task completion.On the other hand,this paper proposes a virtual machine(VM)migration strategy with consideration of multiple mobile users in the MEC network in UDN,and effectively reduces the frequency of VM migration in UDN.Firstly,we propose the wireless link model among the MEC service,which aims at minimizing the energy consumption with considering the mobility of the user and the difference of the link.Secondly,the VM migration algorithm we proposed bases on the combination of energy consumption and computational resources with the consideration of user mobility.Finally,we simulate the model and the result proves the effectiveness of the algorithm.The algorithm can greatly reduce the energy consumption of the whole system and improve the service efficiency of the system.