摘要
多媒体应用要求存储系统提供满足服务质量QoS(QualityofSer vice)需求的数据访问服务 ,反应时间超过截止期限的访问请求必须小于一定的比例 .在开放环境下 ,并发多媒体负载中访问请求的随机性增加了实现这一目标的难度 .并发多媒体应用的访问请求经汇聚后到达存储系统的过程服从泊松分布 ,存储系统的服务时间服从指数分布 ;并发多媒体负载访问存储系统的过程用M/M/1排队论模型描述 .通过对该模型的分析推导出构成并发负载的多媒体应用数量N与负载中访问请求的响应时间分布之间的关系 .仿真试验的结果证明该模型对于保证并发多媒体应用的QoS是非常有效的 .
The multimedia applications need the guarantee of the QoS (quality of service) when retrieving data form the storage systems, the fraction of requests whose response time exceeds a specific delay limit must be below a certain proportion. In the concurrent multimedia workloads, many applications access a storage system at the same time, Which makes it difficult to satisfy the QoS requirements. The aggregate requests of concurrent multimedia applications Poisson arrive the storage system, the storage system service time is exponential distribution, such stochastic process was abstracted as a M/M/1 queueing model. Based on the queueing model, the relationship between the number of concurrent multimedia applications and the statistical distribution of the requests response time was determined, the relationship can be used to guarantee the QoS of concurrent multimedia applications. The experimental results validate that the novel model is accurate.
出处
《北京航空航天大学学报》
EI
CAS
CSCD
北大核心
2004年第11期1043-1047,共5页
Journal of Beijing University of Aeronautics and Astronautics
基金
国家自然科学基金资助项目 (60 2 73 0 3 1)
高等学校博士点专项科研基金资助项目 (2 0 0 2 2 0 0 5 5 0 2 1)
关键词
多媒体负载
存储系统
服务质量
排队模型
Abstracting
Applications
Concurrent engineering
Mathematical models
Multimedia systems
Quality of service
Requirements engineering
Response time (computer systems)
Stochastic programming
