The Content and Access Dynamics of a Busy Web Server

In this paper, we present a detailed study of the dynamics of a busy Web server, at present one of the largest in the Inter-net. We analyze the dynamics of both the server content and the client accesses made to the server. The former includes a characterization of the content creation and modification pro-cess while the latter considers page popularity, and the tem-poral stability and spatial locality in client accesses. Some of our key results are: (a) file modification, though more fre-quent than file creation, often results in little change in the file being modified, (b) a small set of files tends to get modified repeatedly, (c) file popularity follows a Zipf-like distribution with a parameter that is much larger than reported in pre-vious, proxy-based studies, and (d) there is significant tem-poral stability in file popularity but not much stability in the domains from which clients access the popular content. We point out several broad implications of these findings.