一种应用敏感的Web服务请求调度策略

An Application-Aware Web Service Requests Scheduling Strategy

在当前企业级Web服务应用中,性能问题一直是人们关注的重点．然而作为支撑Web服务应用的主流平台,Web应用服务器对请求的调度仍然是传统的先来先服务策略(FCFS)．这种策略无法区分请求的重要性,降低了关键请求的性能．以往的研究较少从应用的性能需求出发考虑服务器的请求调度机制,影响了服务器性能保障的效果．对此,提出了应用敏感的Web服务请求调度策略(AWSRS),使用应用获益来评估服务器为应用提供的性能保障效果．服务器将请求按照应用的需求进行分类,并按照应用获益最大化的目标为不同类型的请求分配资源．实验表明AWSRS策略能够有效提高关键请求的性能．

For commercial Web-services based applications （WSBAs）, performance problem is a major concern. As the main platform for WSBAs, Web application servers （WASs） typically adopt the first-come first-served（FCFS） scheduling strategy. FCFS strategy does not distinguish critical requests from non-critical ones, and therefore the performance of critical ones might be reduced. Most of the existing research seldom takes into account the requirements from applications, which impairs the performance guarantee from WASs. This paper proposes a novel application-aware Web service requests scheduling （AWSRS） strategy for WSBAs, introduces the concept of ＂application benefit＂ to evaluate the effect of WAS in improving the performance for applications. Based on that, the strategy is able to classify requests and intelligently assign corresponding resources for them, in order to maximize the total application benefits. Later experimental results indicate that the strategy effectively improves the performance of critical requests.