摘要
在云环境下,用户能够方便地按需获取满足其要求的服务资源,而高质量的服务将受到用户的青睐.然而,云环境下的服务请求具有即时、并发以及大规模等特点,多个用户可能同时调用同一个高质量的服务,极有可能出现用户访问量超过服务的负载容量的情况,致使服务负载失衡,从而导致无法即时响应用户请求,服务提供能力急剧下降.针对这一问题,本文提出了面向云环境的一种负载感知的服务选择方法.该方法,首先构建了一个描述用户请求和服务QoS之间关系的用户满意度模型.其次,由服务的负载能力和用户的满意度获得服务的综合得分,并根据服务的综合得分对服务进行排名与选择.最后,在数据集上进行了大规模的实验,实验结果表明:负载感知的服务选择方法能在保证负载平衡的情况下有效地帮助用户选择满足需求的高质量的服务.
In a cloud environment, it is desired to select qualified services that meet users ＇ requirements, and the services with expected high QoS are usually preferred by users. However, service requests in a cloud environment are typically real-time, concurrent and largescale. It is probable that load of services will become unbalanced while many users are invoking the same service, and so user＇ s requests aren＇ t satisfied and service provision capability will decline rapidly. To solve this problem, we present a load-aware service selection method in cloud environment considering load balance of services. In this approach, we first build a user satisfaction model which describes the relationship between user＇ s request and service＇ s QoS. Then, we integrate a service＇ s workload with its satisfaction score to achieve composite score of the service, and rank and select services with their composite scores. Finally, by conducting large-scale experiments based on a Web service dataset, we show that load-aware service selection approach can effectively help user select high qualified services while keeping load balance of services.
出处
《小型微型计算机系统》
CSCD
北大核心
2014年第9期1994-1998,共5页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61272063
61100054
61300129)资助
教育部新世纪优秀人才支持计划项目(NCET-10-0140)资助
湖南省杰出青年基金项目(11JJ1011)资助
湖南省自然科学基金项目(12JJB009
12JJ6064)资助
湖南省科技计划项目(2013FJ300)资助
湖南科技大学博士科研基金项目(E51368)资助
湖南省高校创新平台开放基金项目(09K085
12K105)资助
湖南科技大学研究生创新基金项目(S120024)资助
