Abstract

Despite the incredible market penetration of smartphones and exponential growth of the app market, utility of smartphones has been and will remain severely limited by the battery life. As such, energy has increasingly become the scarcest resource on smartphones that critically affects user experience. In this talk, I will start with the first survey study that characterizes smartphone energy bugs, or ebugs, broadly defined as errors in the system (apps, OS, hardware, firmware, or external conditions) that result in unexpected smartphone battery drainage and leads to significant user frustrations.

As a first step towards taming ebugs, we built the first fine-grained energy profiler, eprof, that performs energy accounting and hence answers the very question “where was the energy spent in the app?” at the per-routine, per-thread, and per-process granularity. Building eprof in turn requires developing a fine-grained, online power model that captures the unique asynchronous power behavior of modern smartphones. Using eprof, we dissected the energy drain of several popular apps in Android Market and discovered ebugs in popular apps like Facebook.

While essential, eprof only provides a semi-automatic tool for energy debugging. The “holy grail” in energy debugging in smartphones is to develop fully automatic debugging techniques and tools, which can draw synergies from many areas of computer science including OS, PL, compilers, machine learning, and HCI . I will present the first automatic detection technique based on static compiler analysis for detecting the class of “no-sleep” energy bugs.