ABSTRACT
Smartphones are now equipped with high-performance processors to meet the increasing complexity of apps. However, these processors drain the battery quickly, which has become a major concern for Smartphone users. The latest big.LITTLE multicore architecture provides new energy-saving facilities in addition to traditional DVFS technique. But this mechanism has not been well exploited by current Smartphones. In this paper, we propose a framework that records the performance demands for repetitive app usage scenarios, and adjusts the big.LITTLE processor to an energy-efficient configuration for subsequent executions of these scenarios. We implemented our framework on Android Smartphones, and the experiments achieve 30% energy saving on average without compromising user experience.
- Linux load balancing. http://rtcc.hanyang.ac.kr/rtccw/?page_id=1745.Google Scholar
- Monkeyrunner. http://developer.android.com/tools/help/MonkeyRunner.html.Google Scholar
- Monsoon power monitor. https://www.msoon.com/.Google Scholar
- Xposed. https://github.com/rovo89/XposedBridge.Google Scholar
- Y.-M. Chang et al. A resource-driven dvfs scheme for smart handheld devices. ACM Trans. Embed. Comput. Syst., 2013. Google ScholarDigital Library
- C. Gao et al. A study of mobile device utilization. In ISPASS, 2015 IEEE International Symposium on.Google Scholar
- Gaspar et al. A framework for application-guided task management on heterogeneous embedded systems. ACM Transactions on Architecture and Code Optimization, 2016. Google ScholarDigital Library
- M. Hashemi et al. Efficient execution of bursty applications. IEEE Computer Architecture Letters, 2015. Google ScholarDigital Library
- D. Kim et al. Content-centric display energy management for mobile devices. DAC '14. ACM. Google ScholarDigital Library
- S. Kim et al. An event-driven power management scheme for mobile consumer electronics. IEEE Transactions on Consumer Electronics, 2013.Google Scholar
- X. Li et al. Smartcap: User experience-oriented power adaptation for smartphone's application processor. DATE '13. EDA Consortium. Google ScholarDigital Library
- X. Li et al. Usage history-directed power management for smartphones. In International Conference on Algorithms and Architectures for Parallel Processing. Springer, 2015.Google ScholarCross Ref
- W. Seo et al. Big or little: A study of mobile interactive applications on an asymmetric multi-core platform. In IISWC, 2015 IEEE International Symposium on. Google ScholarDigital Library
- W. Song et al. Reducing energy consumption of smartphones using user-perceived response time analysis. HotMobile '14. ACM. Google ScholarDigital Library
- P.-H. Tseng et al. User-centric energy-efficient scheduling on multi-core mobile devices. DAC '14. ACM. Google ScholarDigital Library
- Y.-H. Wei et al. Energy-efficient real-time scheduling of multimedia tasks on multi-core processors. SAC '10. ACM. Google ScholarDigital Library
- L. Yang et al. Happe: Human and application-driven frequency scaling for processor power efficiency. IEEE Transactions on Mobile Computing, 2013. Google ScholarDigital Library
- L. Zhang et al. Panappticon: Event-based tracing to measure mobile application and platform performance. In CODES+ISSS, 2013 International Conference on. Google ScholarDigital Library
- Y. Zhu et al. Event-based scheduling for energy-efficient qos (eqos) in mobile web applications. In HPCA, 2015.Google ScholarCross Ref
Recommendations
User-Centric Scheduling and Governing on Mobile Devices with big.LITTLE Processors
Mobile applications will become progressively more complicated and diverse. Heterogeneous computing architectures like big.LITTLE are a hardware solution that allows mobile devices to combine computing performance and energy efficiency. However, ...
Performance and energy evaluation of CoMD on Intel Xeon Phi co-processors
Co-HPC '14: Proceedings of the 1st International Workshop on Hardware-Software Co-Design for High Performance ComputingMolecular dynamics simulations are used extensively in science and engineering. Co-Design Molecular Dynamics (CoMD) is a proxy application that reflects the workload characteristics of production molecular dynamics software. In particular, CoMD is ...
Energy-efficient scheduling of a real-time task on DVFS-enabled multi-cores
ICHIT '09: Proceedings of the 2009 International Conference on Hybrid Information TechnologyWe propose an energy-efficient scheduling of a long-lived real-time video task running on DVFS-enabled multi-core platforms. The proposed scheme reduces the energy consumption by executing the task in parallel on an appropriate number of cores with the ...
Comments