面向复杂数据应用的可扩展负载平衡中间件研究

Research on Scalable Load-Balancing Middleware for Complex Data Applications

现有的负载平衡中间件大都采用单一的基于对象的负载平衡粒度,更多地关注于同一对象组的不同副本之间的平衡,不能满足雷达仿真所涉及的复杂数据应用的要求.基于通用对象请求代理体系结构(CORBA)技术,采用基于移动对象适配器(POA)的负载平衡粒度,提出了一种高可扩展的负载平衡中间件模型,阐述了该模型相关的若干关键技术、解决方法及其实现,包括负载策略的扩展、负载度量的扩展等.实验表明:该模型能够有效降低平均响应时间,并增加总的吞吐量.

Most of the existing load-balancing middleware adopt the single object-based load balancing granularity,and pay more attention to the balance among the different replicas of the same object group,which can not meet the demands of the complex data application involved in radar simulation.Therefore,in terms of the technique of common object request broker architecture（CORBA）,this paper presents a model of highly scalable load-balancing middleware by using the load balancing granularity based on the portable object adapter（POA） and illustrates numerous key techniques,solutions and their implementations related to the proposed model,including the extensions of load strategy and load metrics,etc.Experimental results show that the proposed model can reduce effectively the average response time and increase the total throughout as well.