基于改进遗传算法的负载均衡机制的应用研究

Application of load balancing mechanism based on improved genetic algorithm

针对商品房预售款监管平台与银行系统、商品房网上销售系统、商品房网上合同备案系统之间频繁进行大数据量的数据交换共享,时常出现数据高并发报送造成的服务不能及时响应,带来数据不能实现实时高效的传输和处理。本文提出采用改进的遗传算法与动态的负载均衡机制相结合,引入了Markowitz Mean-Variance模型作为遗传算法的适应度函数,用滑动窗口技术解决了大数据量调度拥塞,再采用基于种群的交流选择方法筛选出了更优的个体,最终设计了高效的负载均衡处理机制解决了上述高并发拥塞问题。经测试表明,文中提出的方案具有一定可行性,能较好的提高高并发情况下数据报接处理的效率。

In view of frequent exchange and sharing of large amount of data between regulatory system of pre-sale of commercial houses and banking system,real estate online sales system and commercial realestate online contract filing system,the services caused by high data concurrency often fail to respond timely. Data can not be real-time and efficient transmission and processing. In this paper,the improved genetic algorithm is combined with the dynamic load balancing mechanism. The Markowitz mean-variance model is introduced as the fitness function of genetic algorithm. The sliding window technique is used to solve the problem of large data volume scheduling congestion. The method screened out better individuals,and finally designed an efficient load balancing mechanism to solve the above high concurrency congestion problem. The test shows that the proposed scheme is feasible and can effectively improve the efficiency of datagram processing in high concurrency situations.