Research On Integration Of Micro Datacenter And Passive Optical Network

With the rapid development of IoT and Internet applications,the number of terminals which access to network is increasing rapidly.At the same time,with the progress of technology,the number of business that a single terminal can support and traffic requirements are also increasing.Both of them are resulting an enormous increasing of data which transmitted in the network.In order to cope with the data increasing of the transmission network,deploying the computing and storage devices at the edge of network to constructing local micro data center,filtering out redundant data and reduce the amount of data transmission in the backbone,is a positive attempt in the scheme.In view of this scheme,there are many related researches in recent years,such as fog calculation,CORD,etc.This paper mainly research on integration of micro data center and optical access network,and in order to realize the highest overall satisfaction in the process of deployment,which according to user needs and specific application scenarios.This paper firstly summarizes some basic concepts and related research that needed in the research process,including micro data center and fog computing,optical access network,several ways of micro data center and optical access network integration,the concept of QoE and evaluation methods,and the basic algorithm of resource scheduling and allocation in cloud computing.Then,according to the architecture of the optical access network,this paper describes the business process after the micro data center deployment.Based on this process,a quantitative QoE evaluation function is designed.For the user service,considering about delay and consumptionof the system,using numerical energy normalization to make the data normalization,and we get the evaluation function of QoE system finally.The calculation method of delay and cost is explained in detail,and the calculation process of QoE evaluation function is illustrated in detail by simulation.When the QoE evaluation function takes the optimal value,it is the best resource allocation scheme for the server.Considering the diversity of user traffic types,then,we studies the resource allocation scheme under the condition of multi service coexistence.Based on the consumption of business local process,we determined the priority ordering when services data arrived.Then,the calculation methods of cost and delay are described in detail when multi service coexistence,a quantized QoE model for multi service coexistence is given at last.There are some enlightenments of server configuration by this paper.First,by means of the utility function,we used numerical normalization method to make a evaluate function that contains a variety of environmental factors,including delay,cost,etc.The second is the details of calculation method about average response delay and equipment investment cost.The third is the calculation of the statistical data needed in the offline calculation.And through this study,considering the delay and cost,we can find a balance between user experience and investment in micro data center deployment.