Reducing Energy Consumption and Latency of Applications on Wireless Devices
Date
2018-03-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Energy consumption and delay are the key factors influencing users' quality of experience for applications running on wireless devices. In cellular networks, providing a satisfactory user experience faces several challenges caused by the poor interactions between multiple factors. First, cellular networks suffer from long round trip times and also employ a radio resource control protocol, which keeps the radio of the device in high power state even after completion of a data transfer. Second, in most applications, completing a user action involves many round trips over the high-latency cellular link, which can lead to poor application performance. Moreover, periodic and intermittent traffic pattern exhibited by the majority of applications can result in serious energy inefficiencies. In this thesis, we address these challenges from the network and end device perspectives. First, we design a network-centric solution, called WebPro, that adopts speculative loading and bundling techniques to reduce latency and energy consumption of mobile web browsing. Performance evaluation results obtained through live experiments indicate that WebPro outperforms state-of-the-art, though the degree of improvement varies for different webpages. Then, we focus on energy-delay tradeoff on end devices and design algorithms to balance the energy-delay tradeoff inherent in bundling. Specifically, we formulate a generalized notion of bundling as an online optimization problem. The objective of this problem is to minimize the bundling cost defined as a weighted summation of energy and delay costs. Based on two different energy cost models associated with smartphones and internet of things (IoT) devices, we develop online algorithms to solve the optimization problem. A distinctive feature of our online algorithms is that they do not rely on any assumption about the traffic pattern or nature of applications. We provide theoretical performance bounds for our proposed algorithms by comparing them to an optimal offline algorithm. We evaluate the performance of our algorithms in a range of realistic scenarios using both model-driven simulations and real experiments on a smartphone. Our results show that depending on the delay tolerance level of a user, energy savings ranging from zero to about 100% can be achieved using our algorithms.
Description
Keywords
Computer Networks, Internet of Things, Energy-Delay Tradeoff, Online Algorithms, Browsing Delay, Request Bundling, Smartphones
Citation
Sehati, A. (2018). Reducing Energy Consumption and Latency of Applications on Wireless Devices (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/13062