Toward a Push-Scalable Global Internet

TL;DR

This paper analyzes the scalability issues of push message delivery on the web and proposes a machine learning-based content optimization method to reduce connection times while maintaining real-time delivery for most messages.

Contribution

It introduces a content-based optimization approach using machine learning to reduce always-on connections in push messaging systems, improving scalability.

Findings

Active connection hours can be halved without losing real-time delivery for 90% of messages.

Measurement analysis highlights scalability challenges of current push email services.

Machine learning accurately models message arrival patterns.

Abstract

Push message delivery, where a client maintains an ``always-on'' connection with a server in order to be notified of a (asynchronous) message arrival in real-time, is increasingly being used in Internet services. The key message in this paper is that push message delivery on the World Wide Web is not scalable for servers, intermediate network elements, and battery-operated mobile device clients. We present a measurement analysis of a commercially deployed WWW push email service to highlight some of these issues. Next, we suggest content-based optimization to reduce the always-on connection requirement of push messaging. Our idea is based on exploiting the periodic nature of human-to-human messaging. We show how machine learning can accurately model the times of a day or week when messages are least likely to arrive; and turn off always-on connections these times. We apply our approach to a real email data set and our experiments demonstrate that the number of hours of active always-on connections can be cut by half while still achieving real-time message delivery for up to 90% of all messages.