Optimization Design And Implementation Of E-commerce Platform Server Performance

In order to adapt to the technological wave of today's Internet,many small and medium-sized enterprises,especially those in traditional industries,are exploring ways and methods to combine their businesses with the Internet.The most common one is to turn their business from offline to online by the Internet deal,that is,e-commerce transaction.In order to fulfill those small and medium-size enterprises' platform designing and cost control requirements,this paper takes the original platform basis to optimization design,by using the existing mature and stable framework and technology,and basing on micro service system,it create a distributed electric business platform system,which provide a more feasible,relatively higher performance and lower cost solution for small and medium enterprises to build their own electrical business platform.In order to achieve higher performance and higher reliability,the server cluster system is constructed with lower cost and lower performance.To solve the load balancing problem of multiple service cases in the micro-service architecture,this paper proposes a load balancing strategy which is based on improved genetic algorithm.In this paper,the application of genetic algorithm in the field of server cluster load balancing,as well as the solution to avoid the phenomena of premature and convergence of genetic algorithm are improved.It also do the chromosome design and fitness function design for the load balancer.Simulated annealing was used to improve the convergent limitation of traditional selection operation and maintain the diversity of population.The idea of acceptance probability in sampling inspection is applied to crossover operation which makes the crossover result more conducive to the search of the optimal solution and accelerates the convergence of the algorithm.The nonlinear inertia weight is introduced into the variation rate to form the dynamic variation rate.Therefore,the limitation of premature and convergence are improved to some extent.In order to verify the effectiveness of the optimal design of the system architecture in this paper and the proposal of load balancing strategy based on the improved genetic algorithm,an experimental comparison was made between the system with the optimized architecture design and the system built before the optimization under the same environmental condition.The performance test and experimental results are compared and analyzed from two aspects of page access and data access.The results show that the performance of the new system is better than that of the original system.At the same time,the performance and effectiveness of the load balancing strategy based on the improved genetic algorithm which proposed in this paper are also tested by experiments,and compared with the weighted minimum response time algorithm and the standard genetic algorithm.The experiment verifies the effect of this algorithm on improving the load balance of server cluster.