The rapid advent of “Web 2.0” applications has unleashed new HTTP traffic patterns which differ from the conventional HTTP request-response model. In particular, asynchronous pre-fetching of data in order to provide a smooth web browsing experience and richer HTTP payloads (e.g., Javascript libraries) of Web 2.0 applications induce larger, heavier, and more bursty traffic on the underlying networks. We present a traffic study of Web 2.0 applications including Google Maps, modern Web-email, and social networking Web sites, and compare them with all HTTP traffic. We highlight the key differences of Web 2.0 traffic from traditional HTTP traffic through statistical analysis. As such our work elucidates the changing face of one of the most popular application on the Internet: The World Wide Web.

The World Wide Web [1] is one of the most popular applications of the Internet that runs primarily over the HTTP protocol. While HTTP (Hyper Text Transfer Protocol) [2] constitutes the session layer or messaging protocol of the Web, the HTML (Hyper Text Markup Language) describes the content and allows authors to connect up web pages through hypertext links or hyperlinks; an idea made popular by Tim Burners Lee in the early 1990s and widely used today. In its classical form, users reach other pages or access new data by clicking on hyperlinks or submitting Web based forms. In this basic HTTP request-response model each clicked link or submitted form results in loading of a new web page in response to the respective request.

The recent popularity of asynchronouscommunication enabled web sites has caused a fundamental shift in the classical HTTP request-response model of the Web. Wide-spread implementation of this approach is usually executed through AJAX (Asynchronous JavaScript and XML) [3], a compendium of technologies that enable Web browsers to request data from the server asynchronously, i.e., without requiring human intervention such as clicking on a hyperlink or on a button. Consequently, HTTP requests are becoming automated rather than being human-generated.

Download pdf The New Web: Characterizing AJAX Traffic