Loading [MathJax]/extensions/MathMenu.js
Characterizing Mobile Web Traffic: A Case Study of an Academic Web Server | IEEE Conference Publication | IEEE Xplore

Characterizing Mobile Web Traffic: A Case Study of an Academic Web Server


Abstract:

With the proliferation of Internet-based technologies over the past two decades and the associated growth in the volume and diversity of Internet traffic, it becomes incr...Show More

Abstract:

With the proliferation of Internet-based technologies over the past two decades and the associated growth in the volume and diversity of Internet traffic, it becomes increasingly important to understand how these changes affect the overall workload characteristics of servers. This paper revisits the seminal work of Arlitt and Williamson [1] to determine whether or not the ten invariants they derived from server logs continue to adequately characterise modern web traffic. Furthermore, a specialised analysis is performed to determine how well these invariants model mobile web traffic in particular. Our results show that while the majority of the invariants hold, some do not. In particular, combined and mobile web traffic has dramatically changed in the file types requested, and the origin of the hosts making requests, as well as a noticeable change in response types. Furthermore, mobile web traffic demonstrated significantly fewer one-time requests as compared to the original study and the combined logs from this study.
Date of Conference: 04-06 November 2019
Date Added to IEEE Xplore: 24 February 2020
ISBN Information:
Conference Location: Kathmandu, Nepal

Contact IEEE to Subscribe

References

References is not available for this document.