关键词: Web系统优化, 混合二进制粒子群优化, 经验因子, 爬山算法, 线性递减惯性权重

Abstract: With the dramatic increase in the amount of network users, Web server is widely used in large software systems and usually needs to take a lot of effort to configure multiple parameters related with performance before running. The process of manual configuration parameters is too cumbersome and requires professional knowledge and experience. To obtain the reasonable web system configuration parameters conveniently and quickly, an optimization algorithm of web system based on hybrid binary particle swarm optimization algorithm is presented. The algorithm which joins the experience factor, hill-climbing algorithm and the linearly decreasing inertia weight can automatically find the optimal configuration parameter of Web system. The algorithm can solve the problems of traditional Binary Particle Swarm Optimization（BPSO） algorithm which has low efficiency and easy to fall into local optimal solution. Experimental results show that the algorithm has high efficiency, can jump out of local optimal solution, and can get better global optimal solution.

Key words: optimization of Web system, hybrid binary particle swarm optimization, experience factor, mountain climbing algorithm, linearly decreasing inertia weight