作者: Alec Wolman , M. Voelker , Nitin Sharma , Neal Cardwell , Anna Karlin
关键词: Computer science 、 Population 、 Web proxy 、 TRACE (psycholinguistics) 、 Content centric 、 Trace analysis 、 Data science 、 Proxy (statistics) 、 Scale (social sciences) 、 World Wide Web
摘要: While algorithms for cooperative proxy caching have been widely studied, little is understood about cooperative-caching performance in the large-scale World Wide Web environment. This paper uses both trace-based analysis and analytic modelling to show potential advantages drawbacks of inter-proxy cooperation. With our traces, we evaluate quantitatively performance-improvement cooperation between 200 small-organization proxies within a university environment, two large-organization handling 23,000 60,000 clients, respectively. model, extend beyond these populations project behavior regions with millions clients. Overall, demonstrate that has benefits only limited population bounds. We also use model examine implications future trends Web-access traffic.