Redundancy Elimination Might Be Overrated: A Quantitative Study on Real-World Wireless Traffic

TL;DR

This study critically examines the practical potential of redundancy elimination in mobile wireless traffic, revealing that real-world savings are often lower than previously claimed due to user and content diversity.

Contribution

It provides a quantitative analysis showing that actual redundancy savings are limited without synchronized user interests, challenging optimistic prior estimates.

Findings

Redundancy savings are lower than literature suggests.

User and domain diversity reduce potential redundancy.

Video and encryption diminish redundancy elimination effectiveness.

Abstract

With significant increases in mobile device traffic slated for the foreseeable future, numerous technologies must be embraced to satisfy such demand. Notably, one of the more intriguing approaches has been blending on-device caching and device-to-device (D2D) communications. While various past research has pointed to potentially significant gains (30%+) via redundancy elimination (RE), some skepticism has emerged to whether or not such gains are truly harnessable in practice. The premise of this paper is to explore whether or not significant potential for redundancy elimination exists and whether the rise of video and encryption might blunt said efforts. Critically, we find that absent significant synchronized interests of mobile users, the actual redundancy falls well short of the promising values from the literature. In our paper, we investigate the roots for said shortcomings by exploring RE savings with regards to cache hit characteristics and to what extent client and domain diversity contribute to the realized redundancy savings.