Web browser workload characterization for power management on HMP platforms

Abstract

The volume of mobile web browsing traffic has significantly increased as well as the complexity of the mobile websites mandating high-performance web page rendering engines to be used on mobile devices. Although there has been a significant improvement in performance of web page rendering on mobile phones in recent years, the power consumption reduction has not been addressed much. A main contribution of this work is a thread level analysis of the workload generated by Google's Chrome browser on a heterogeneous multiprocessing (HMP) platform found in many smartphones. We analyze the detailed traces of the thread workload generated by the web browser, especially the rendering engine, and discuss the power saving potentials in relation to power management policies in Android. Moreover, we propose power management strategies based on the results. All trace data and measurement results have been collected on a real HMP platform integrating the Samsung Exynos5422 SoC, also used in the Samsung Galaxy S5 smartphone. Our work shows that there is a considerable scope for power savings and outlines directions for future research. We believe that it will lead to development of practical power management techniques considering thread allocation, dynamic voltage and frequency scaling (DVFS) and power gating.