针对在军事信息栅格体系内, 服务聚合后的部署状态对网络节点性能和服务质量(QoS)产生反馈影响的问题, 提出了一种带有负载均衡的服务聚合模型(load balancing composition model for military information service, LBCMMS), 通过重新构建QoS参数组, 使得对服务的优选评估更符合军事需求; 引入动态削弱矩阵(dynamic weakening watrix, DWM), 并定义矩阵中的两种负指数削弱因子, 促使聚合的服务流在节点的分布上具有负载均衡效果; 分别基于回溯—贪婪算法(BGA)和遗传算法(GA)设计了服务聚合流程, 解决了模型多条件决策下的NP-complete问题; 通过仿真验证了模型的有效性。
In system of military information grid, disposition state after service compositing has feedback influence on perfor-mance of network node and QoS of service cluster. Due to this problem, this paper proposed a new model LBCMMS. By reseting QoS parameters, assessment of service got more fitter for military demands. It introduced DWM, and defined two kinds of negative exponent weakening factors in it, which would impel disposition state of services has effect of load balancing. Based on BGA and GA, designed the services compositing process, which solved the NP-complete problem of decision-making with multi-condition in this model. Through simulation experiments, effectiveness of the model is verified.