安卓系统为浏览器分配资源时无法感知网页内容，会导致资源过度分配和电量不必要损失。同时，由于CPU可调节频率密度的增长，通过动态电压频率缩放（dynamic voltageand frequency scaling,DVFS）技术实现能耗优化的难度也随之增大。另外在系统默认的调控策略下，忽视了图形处理器（graphics processing unit,GPU）对浏览器运行的作用。针对上述问题，提出一种协同调控CPU和GPU实现功耗优化的方法。首先根据网页加载时处理器运行特征利用逻辑回归对网页进行分类，对网页特征加权实现复杂度量化，根据类别与复杂度采用DVFS技术限制CPU频率的同时调节GPU频率。该方法被应用于谷歌Pixel2 XL上的Chromium浏览器，对排名前500的中文网站进行测试，平均节省了12%功耗的同时减少了5%网页加载时间。
Android's inability to sense web page content during resources allocation to the browser often results in over-allocation of resources and unnecessary loss of power. At the same time, due to the growth of CPU adjustable frequency density, optimizing energy consumption through dynamic voltage and frequency scaling (DVFS) technology becomes increasingly challenging. Furthermore, the role of the graphics processing unit (GPU) in browser operation is ignored under the system's default regulation policy. Aiming at the above problems, we propose a method to optimize power consumption by co-regulating CPU and GPU. First, web pages are classified by logistic regression based on the processor operating characteristics when loading web pages. We assign weights to webpage characteristics to quantify the complexity, and then use DVFS to limit the CPU frequency while adjusting the GPU frequency based on webpage category and complexity. The proposed method is applied to the Chromium browser on Google Pixel2 XL, and tested on the top 500 Chinese websites, resulting in a 12% reduction in power consumption and an average 5% decrease in webpage loading time.
[1] 王魁祎, 周改云. Android平台的移动APP开发策略研究[J]. 软件, 2021, 42(4): 144-146. Wang K Y, Zhou G Y. Research on mobile APP development strategy based on android platform [J]. Software, 2021, 42(4): 144-146. (in Chinese)
[2] 石烺峰. APP应用程序开发模式探究[J]. 电子制作, 2019(19): 95-96, 59 Shi L F. Research on APP application development mode [J]. Practical Electronics, 2019(19): 95-96, 59. (in Chinese)
[3] Zhu Y H, Reddi V J. High-performance and energy-efficient mobile web browsing on big/little systems [C]//IEEE 19th International Symposium on High Performance Computer Architecture. IEEE, 2013: 13-24.
[4] Dambrosio S, De Pasquale S, Iannone G, et al. Energy consumption and privacy in mobile web browsing: individual issues and connected solutions [J]. Sustainable Computing: Informatics and Systems, 2016, 11: 63-79.
[5] Thiagarajan N, Aggarwal G, Nicoara A, et al. Who killed my battery?: analyzing mobile browser energy consumption [C]//The 21st International Conference on World Wide Web, 2012: 41-50.
[6] Peters N, Park S, Clifford D, et al. Phase-aware web browser power management on hmp platforms [C]//International Conference on Supercomputing, 2018: 274-283.
[7] Zhu Y H, Halpern M, Reddi V J. Event-based scheduling for energy-efficient QoS (eQoS) in mobile web applications [C]//IEEE 21st International Symposium on High Performance Computer Architecture (HPCA), 2015: 137-149.
[8] Zhu Y H, Halpern M, Reddi V J. The role of the CPU in energy-efficient mobile web browsing [J]. IEEE Micro, 2015, 35(1): 26-33.
[9] Choi Y, Park S, Cha H. Optimizing energy efficiency of browsers in energy-aware schedulingenabled mobile devices [C]//The 25th Annual International Conference on Mobile Computing and Networking, 2019: 1-16.
[10] Yuan L, Ren J, Gao L, et al. Using machine learning to optimize web interactions on heterogeneous mobile systems [J]. IEEE Access, 2019, 7: 139394-139408.
[11] Meyerovich L A, Bodik R. Fast and parallel webpage layout [C]//The 19th International Conference on World Wide Web, 2010: 711-720.
[12] Sivakumar A, Narayanan S P, Gopalakrishnan V, et al. PARCEL: proxy assisted browsing in cellular networks for energy and latency reduction [C]//The 10th ACM International on Conference on Emerging Network Experiment and Technology, 2014: 325-336.
[13] Zhao B, Hu W J, Zheng Q, et al. Energy-aware web browsing on smartphones [J]. IEEE Transactions on Parallel and Distributed Systems, 2015, 26(3): 761-774.
[14] Peters N, Park S, Chakraborty S, et al. Web browser workload characterization for power management on HMP platforms [C]//The Eleventh IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis, 2016: 1-10.
[15] Xu F, Yang S, Zhou Z, et al. eBrowser: making human-mobile web interactions energy efficient with event rate learning [C]//IEEE 38th International Conference on Distributed Computing Systems (ICDCS), 2018: 523-533.
[16] Kim D, Ko Y B, Lim S H. Energy-efficient real-time multi-core assignment scheme for asymmetric multi-core mobile devices [J]. IEEE Access, 2020, 8: 117324-117334.
[17] Ren J, Wang X M, Fang J B, et al. Proteus: network-aware web browsing on heterogeneous mobile systems [C]//The 14th International Conference on Emerging Networking Experiments and Technologies, 2018: 379-392
[18] Bui D H, Liu Y X, Kim H, et al. Rethinking energy-performance trade-off in mobile web page loading [C]//The 21st Annual International Conference on Mobile Computing and Networking, 2015: 14-26.
[19] Kim S, Bin K, Ha S, et al. zTT: learning-based DVFS with zero thermal throttling for mobile devices [C]//The 19th Annual International Conference on Mobile Systems, Applications, and Services, 2021: 41-53.
[20] Li X L, Yan G H, Han Y H, et al. SmartCap: using machine learning for power adaptation of smartphone's application processor [J]. ACM Transactions on Design Automation of Electronic Systems, 2014, 20(1): 1-16.
[21] 高岭, 任杰, 王海, 等. 基于支持向量机的移动Web浏览性能优化研究[J]. 计算机学报, 2018, 41(9): 2077-2088. Gao L, Ren J, Wang H, et al. Optimize mobile web browsing based on support vector machine [J]. Chinese Journal of Computers, 2018, 41(9): 2077-2088. (in Chinese)
