Study On Grey Wolf Optimization Algorithm In Internet Service Scene

With the rapid development of the Internet and cloud computing technology,the surge of mobile users has led to the rapid growth of Internet traffic,and a large number of users have issued applications for the use of computing resources.How to make efficient and accurate prediction and analysis of mobile Internet traffic data,so as to provide better mobile network services for users;Under the condition of limited resources,how to satisfy the reasonable allocation of virtual machine requests and maximize the profit of cloud service providers are the most concerned problems of the suppliers.For internet service background,this paper studies the improved grey wolves optimization algorithm and is used to solve the problem of internet traffic prediction and cloud service virtual machine profit optimization of resource distribution,gaussian wolves optimization algorithm is proposed and two kinds of discrete grey wolf optimization algorithm named the improved algorithm,the traffic forecasting model's parameters were optimized with GTGWO,virtual machine allocation of resources to solve by the SGWO profit maximization model,the relevant work is as follows:(1)Research and application of gaussian grey wolf optimization algorithmFor the initial population of grey wolf algorithm,gaussian distribution function is introduced to improve the distribution structure of the initial population.Aiming at the convergence mode of the grey wolf algorithm,the gaussian grey wolf optimization algorithm GTGWO was proposed by improving the convergence factor to speed up the optimization process and improve the ability of searching the optimal solution of the grey wolf.In order to test the performance of the new algorithm,it is compared with the same type of GWO algorithm and other intelligent algorithms for 10 benchmark functions.The results show that the stability and optimization speed of GTGWO algorithm are the best.Finally,GTGWO is used to optimize the parameters of the internet traffic prediction model,and a higher prediction accuracy is obtained.(2)Research and application of discrete grey wolf optimization algorithmTo solve the profit optimization problem of virtual machine resource allocation,a discrete grey wolf optimization algorithm SGWO based on dynamic weight strategy and the introduction of dynamic constant was proposed.In order to test the performance of the new algorithm,the actual case data of BKP(WBPK,UBPK,SBPK),three types of bounded knapsack problems,are used as test objects.At the same time,compared with the existing intelligent algorithms,the results show that the performance of SGWO algorithm is outstanding,such as effectiveness and stability.Finally,SGWO algorithm is used to solve the optimization model of virtual machine scheduling resource allocation in cloud computing server,which can better solve such complex dynamic knapsack problem.