Event-based scheduling for energy-efficient QoS (eQoS) in mobile Web applications

Abstract

Mobile Web applications have become an integral part of our society. They pose a high demand for application quality of service (QoS). However, the energy-constrained nature of mobile devices makes optimizing for QoS difficult. Prior art on energy efficiency optimizations has only focused on the trade-off between raw performance and energy consumption, ignoring the application QoS characteristics. In this paper, we propose the concept of energy-efficient QoS (eQoS) to capture the trade-off between QoS and energy consumption. Given the fundamental event-driven nature of mobile Web applications, we further propose event-based scheduling as an optimization framework for eQoS. The event-based scheduling automatically reasons about users' QoS requirements, and accurately slacks the events' execution time to save energy without violating end users' experience. We demonstrate a working prototype using the Google Chromium and V8 framework on the Samsung Exynos 5410 SoC (used in the Galaxy S4 smartphone). Based on real hardware and software measurements, we achieve 41.2% energy saving with only 0.4% of QoS violations perceptible to end users. 137978-1-4799-8930-0/15/$31.00 ©2015 IEEE

Efficient Execution of Bursty Applications

“…Chadha et al propose instruction prefetching based on event-signatures [2]. Zhu et al discuss QoS-driven scheduling techniques for event-based applications [13] and hardware and observe that web-page loading is not generally network limited and can benefit from heavyweight compute hardware [14].…”

A study of mobile device utilization

“…Zhu and Reddi [4,26] propose two specialized hardware and an event-based scheduling for mobile web applications. Several papers have presented work on addressing the Dark Silicon problem [27][28][29][30].…”

Phase-Aware Web Browser Power Management on HMP Platforms

“…It uses a predictive model that is trained using web page primitives such as CSS and HTML tags. Another approach profiles user and system events to identify the quality of service (QoS) required by a mobile web application [45]. This data is used to perform CPU task allocation and DVFS on a big.LITTLE platform.…”

“…The main distinguishing feature of our work over the above works is that while they use indirect information of the context -e.g., web page primitives or user events -we make use of a browser's internal information directly, which allows more effective power management. For example, we use states such as Video or Load/Idle, which are not detectable from user events as in [45] (see Section 5).…”

“…While power management of mobile devices has been extensively studied for several years now, the focus has been on video decoding applications [24,38], and also on games [13,25]. Surprisingly, power management techniques specifically targeting the web browser have been less studied [35,45,48]. But both, its importance, and its potential for power savings, is increasingly being recognized.…”