摘要
针对采用服务组合方式实现流媒体业务提供过程中的服务路径选择问题,从物理资源和逻辑资源两个角度出发,选择时延、带宽、丢包率和服务实例可用服务能力作为参数,建立了多约束归一化线性模型;同时,根据模型特点,采用线性整数规划迭代求解,提高了处理效率。仿真实验证明,该模型和算法能够有效地保障业务服务质量,并具备良好的负载均衡效果。
Focus on the scenario that service composition took place to provide stream media service, presented a novel service routing algorithm. It took delay, bandwidth, package loss rate and service access ability as its metrics, and transfered multiple constraints to linear constraints, and then be resolved using linear programming. The simulation results show that it not only solves the problem of quality of service, but also keeps the load balancing among service instances.
出处
《计算机应用研究》
CSCD
北大核心
2009年第10期3867-3870,共4页
Application Research of Computers
基金
国家"863"计划资助项目(2008AA01A317)
关键词
流媒体
服务路径
归一化线性模型
服务质量
负载均衡
stream media
service routing
normalize linear model
quality of service
load balancing
