Web集群服务器中分发静态请求的调度算法

为分发静态请求提出一种自适应的基于文档大小的调度算法ADSB，它使用资源占用时间来衡量负载，并根据被请求文档的大小来分发请求，均衡各个后台服务器的负载；ADSB根据负载历史周期地预测即将到来的负载统计特性，并根据预测结果来调整算法的参数；由于有着目标位置特性，ADSB能获得很高的缓存命中率；因为实际静态文档的大小服从重尾分布，所以分发大小不同的文档到不同的服务器，使ADSB减小了小文档的平均响应时间，同时大文档也没受到明显的影响。实验表明，ADSB的性能优于已有的经典的调度算法。

Scheduling Algorithm for Distributing Static Request in Web Server Cluster

This paper proposes an adaptive document-size-based scheduling algorithm for distributing static request called ADSB.It also proposes to use occupying resource time to measure load.The dispatcher of ADSB distributes static requests according to the request content size and tries to balance the load among real servers.ADSB periodically predicts the statistical characteristics of future workload based on workload history,and then adjusts system parameters according to the prediction.Because of its locality-aware nature,ADSB achieves very high cache hit ratio.For the sizes of realistic static Web documents follow heavy-tailed distribution,distributing "long" and "short" jobs to different real servers makes ADSB reduce the average job slowdown.Experiments show that ADSB,as compared with some existing classic scheduling algorithms,achieves superior performance.