``Does Internet media traffic really follow Zipf-like distribution?" 
 
Lei Guo, Enhua Tan, Songqing Chen, Zhen Xiao, and Xiaodong Zhang

Proceedings of ACM SIGMETRICS'07 Conference, (Extended Abstract), 
San Diego, California, June 12-16, 2007. 

Abstract

It is commonly agreed that Web traffic follows the Zipf-like distribution, 
which is an analytical foundation for improving Web access performance 
by client-server based proxy caching systems on Internet. In this study, 
we have analyzed a wide variety of media workloads on the Internet, and found 
that the reference ranks of media objects in all the workloads follow 
"stretched exponential" (SE) distribution. This model implies that P2P 
collaborative caching systems can effectively deliver Internet media content. 
Current technology advancements, such as PPLive and BitTorrent have 
demonstrated the strong advantages of P2P collaboration on the delivery 
of Internet media content.