An Integrated Energy Management Framework for Multiple Side-by-Side Applications on Smartphones

While simultaneous side-by-side execution of multiple applications in split-screen mode on smartphones improves quality-of-service (QoS) for end-users, it also results in increased power consumption and reduced battery lifetime. Saving energy when multiple concurrent applications run side-by-side is extremely challenging since applications often utilize the same shared system resources, e.g., memory, at the same time, and resource needs change over time. We present an application-aware integrated system-level energy management framework that (1) leverages applications' offline profiles to detect dynamic changes in resource usage patterns of the workload of applications, and (2) opportunistically throttle (and later redeem) applications based on their instantaneous resource usage characteristics by dynamically adjusting both the voltage and frequency settings of the processor and memory bandwidth at runtime to achieve the most energy-efficient configuration subject to QoS constraints. Experiments on a Pixel 2 smartphone show that our approach achieves an average energy reduction of 13% (16%) and up to 16% (20%) compared to the most closely related work [1] (and default Android governor) for different combinations of real-world applications running side-by-side in split-screen mode. Our approach is also able to reduce the energy consumption of newly installed real-world applications for which there exists no prior resource usage data by up to 12% (14%) when compared to [1] (and default Android governor).