{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,16]],"date-time":"2026-01-16T03:15:59Z","timestamp":1768533359548,"version":"3.49.0"},"reference-count":61,"publisher":"Association for Computing Machinery (ACM)","issue":"1","license":[{"start":{"date-parts":[[2025,3,20]],"date-time":"2025-03-20T00:00:00Z","timestamp":1742428800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Key-Area R&D Program of Guangdong","award":["2021B0101310002"],"award-info":[{"award-number":["2021B0101310002"]}]},{"DOI":"10.13039\/501100001809","name":"NSFC","doi-asserted-by":"crossref","award":["62072432"],"award-info":[{"award-number":["62072432"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Archit. Code Optim."],"published-print":{"date-parts":[[2025,3,31]]},"abstract":"<jats:p>Mobile devices need to respond quickly to diverse user inputs. The existing approaches often heuristically raise the CPU\/GPU frequency according to the empirical rules when facing burst inputs and various changes. Although doing so can be effective sometimes, the existing approaches still need improvements. For instance, raising processors\u2019 frequency can lead to high power consumption when the frequency is over-provisioned or fail to meet user demands when the frequency is under-provisioned. To this end, we propose MobiRL, a reinforcement learning-based scheduler for intelligently adjusting the CPU\/GPU frequency to satisfy user demands accurately on mobile systems. MobiRL monitors the mobile system status and autonomously learns to optimize UI smoothness and power consumption by conducting CPU\/GPU frequency-adjusting actions. The experimental results on the latest delivered smartphones show that MobiRL outperforms the widely used commercial scheduler on real devices\u2014reducing the frame drop rate by 4.1% and reducing power consumption by 42.8%, respectively. Moreover, compared with a study using Q-Learning for CPU frequency scheduling, MobiRL achieves up to a 2.5% lower frame drop rate and reduces power consumption by 32.6%, respectively. Our approach has been deployed in mobile phone products.<\/jats:p>","DOI":"10.1145\/3674910","type":"journal-article","created":{"date-parts":[[2024,10,26]],"date-time":"2024-10-26T09:27:28Z","timestamp":1729934848000},"page":"1-27","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["An Intelligent Scheduling Approach on Mobile OS for Optimizing UI Smoothness and Power"],"prefix":"10.1145","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0009-0008-8828-1404","authenticated-orcid":false,"given":"Xinglei","family":"Dou","sequence":"first","affiliation":[{"name":"Beihang University, Beijing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4854-7382","authenticated-orcid":false,"given":"Lei","family":"Liu","sequence":"additional","affiliation":[{"name":"Beihang University, Beijing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9438-9181","authenticated-orcid":false,"given":"Limin","family":"Xiao","sequence":"additional","affiliation":[{"name":"Beihang University, Beijing, China"}]}],"member":"320","published-online":{"date-parts":[[2025,3,20]]},"reference":[{"key":"e_1_3_2_2_2","unstructured":"F. Laricchia. Market share of mobile operating systems worldwide 2012-2022. https:\/\/www.statista.com\/statistics\/272698\/global-market-share-held-by-mobile-operating-systems-since-2009\/"},{"key":"e_1_3_2_3_2","unstructured":"R. J. Wysocki. CPU performance scaling. https:\/\/docs.kernel.org\/admin-guide\/pm\/cpufreq.html"},{"key":"e_1_3_2_4_2","unstructured":"Device frequency scaling\u2014the Linux kernel documentation. https:\/\/docs.kernel.org\/driver-api\/devfreq.html"},{"key":"e_1_3_2_5_2","unstructured":"Google. power\/1.0 - platform\/hardware\/interfaces - Git at Google. https:\/\/android.googlesource.com\/platform\/hardware\/interfaces\/+\/master\/power\/1.0\/default\/Power.h"},{"key":"e_1_3_2_6_2","unstructured":"Samsung. Gamedev | Samsung developers. https:\/\/developer.samsung.com\/galaxy-gamedev\/overview.html"},{"key":"e_1_3_2_7_2","unstructured":"OPPO. Hyper boost | OPPO developer. https:\/\/developers.oppomobile.com\/newservice\/capability?pagename=hyper_boost"},{"key":"e_1_3_2_8_2","unstructured":"Huawei. Service Introduction-PerfGenius-Accelerate Kit | HUAWEI developers. https:\/\/developer.huawei.com\/consumer\/en\/doc\/development\/HMSCore-Guides-V5\/introduction-0000001054817121-V5"},{"key":"e_1_3_2_9_2","doi-asserted-by":"crossref","unstructured":"N. Mishra C. Imes J. D. Lafferty and H. Hoffmann. 2018. CALOREE: learning control for predictable latency and low energy. In ASPLOS.","DOI":"10.1145\/3173162.3173184"},{"key":"e_1_3_2_10_2","volume-title":"FIRM: An intelligent fine-grained resource management framework for SLO-oriented microservices. In OSDI.","author":"Qiu H.","year":"2020","unstructured":"H. Qiu, S. S. Banerjee, S. Jha, Z. T. Kalbarczyk, and R. K. Iyer. 2020. FIRM: An intelligent fine-grained resource management framework for SLO-oriented microservices. In OSDI."},{"key":"e_1_3_2_11_2","volume-title":"Pythia: A customizable hardware prefetching framework using online reinforcement learning. In MICRO.","author":"Bera R.","year":"2021","unstructured":"R. Bera, K. Kanellopoulos, A. Nori, T. Shahroodi, S. Subramoney, and O. Mutlu. 2021. Pythia: A customizable hardware prefetching framework using online reinforcement learning. In MICRO."},{"key":"e_1_3_2_12_2","volume-title":"Ml-gov: A machine learning enhanced integrated CPU-GPU DVFS governor for mobile gaming. In ESTIMedia.","author":"Park J.","year":"2017","unstructured":"J. Park, N. D. Dutt, and S. Lim. 2017. Ml-gov: A machine learning enhanced integrated CPU-GPU DVFS governor for mobile gaming. In ESTIMedia."},{"key":"e_1_3_2_13_2","unstructured":"Qualcomm. Qualcomm Snapdragon 888 5g mobile platform | Qualcomm. https:\/\/www.qualcomm.com\/products\/application\/smartphones\/snapdragon-8-series-mobile-platforms\/snapdragon-888-5g-mobile-platform"},{"key":"e_1_3_2_14_2","doi-asserted-by":"crossref","unstructured":"S. J. Tarsa R. B. R. Chowdhury J. Sebot G. N. Chinya J. Gaur K. Sankaranarayanan C. Lin R. Chappell R. Singhal and H. Wang. 2019. Post-silicon CPU adaptation made practical using machine learning. In ISCA.","DOI":"10.1145\/3307650.3322267"},{"key":"e_1_3_2_15_2","doi-asserted-by":"crossref","unstructured":"J. Rahman and P. Lama. 2019. Predicting the end-to-end tail latency of containerized microservices in the cloud. In IC2E.","DOI":"10.1109\/IC2E.2019.00034"},{"key":"e_1_3_2_16_2","volume-title":"APOLLO: An automated power modeling framework for runtime power introspection in high-volume commercial microprocessors. In MICRO.","author":"Xie Z.","year":"2021","unstructured":"Z. Xie, X. Xu, M. Walker, J. Knebel, K. Palaniswamy, N. Hebert, J. Hu, H. Yang, Y. Chen, and S. Das. 2021. APOLLO: An automated power modeling framework for runtime power introspection in high-volume commercial microprocessors. In MICRO."},{"key":"e_1_3_2_17_2","doi-asserted-by":"crossref","unstructured":"J. Tan T. Zhang F. Li J. Chen Q. Zheng P. Zhang H. Qiao Y. Shi W. Cao and R. Zhang. 2019. iBTune: Individualized buffer tuning for large-scale cloud databases. In VLDB Endowment.","DOI":"10.14778\/3339490.3339503"},{"key":"e_1_3_2_18_2","doi-asserted-by":"crossref","unstructured":"J. Chen S. S. Banerjee Z. T. Kalbarczyk and R. K. Iyer. 2020. Machine learning for load balancing in the Linux kernel. In APSys.","DOI":"10.1145\/3409963.3410492"},{"key":"e_1_3_2_19_2","volume-title":"Twig: Multi-agent task management for colocated latency-critical cloud services. In HPCA.","author":"Nishtala R.","year":"2020","unstructured":"R. Nishtala, V. Petrucci, P. M. Carpenter, and M. Sj\u00e4lander. 2020. Twig: Multi-agent task management for colocated latency-critical cloud services. In HPCA."},{"key":"e_1_3_2_20_2","doi-asserted-by":"crossref","unstructured":"E. Ipek O. Mutlu J. F. Mart\u00ednez and R. Caruana. 2008. Self-optimizing memory controllers: A reinforcement learning approach. In ISCA.","DOI":"10.1109\/ISCA.2008.21"},{"key":"e_1_3_2_21_2","unstructured":"T. P. Lillicrap J. J. Hunt A. Pritzel N. Heess T. Erez Y. Tassa D. Silver and D. Wierstra. 2016. Continuous control with deep reinforcement learning. In ICLR."},{"key":"e_1_3_2_22_2","unstructured":"On-device training with Tensorflow lite. https:\/\/www.tensorflow.org\/lite\/examples\/on_device_training\/overview"},{"key":"e_1_3_2_23_2","unstructured":"Analyze with Profile GPU Rendering | App quality | Android developers. https:\/\/developer.android.com\/topic\/performance\/rendering\/profile-gpu"},{"key":"e_1_3_2_24_2","unstructured":"kernel\/sched\/boost.c - kernel\/msm - Git at Google. https:\/\/android.googlesource.com\/kernel\/msm\/+\/refs\/tags\/android-10.0.0_r0.4\/kernel\/sched\/boost.c"},{"key":"e_1_3_2_25_2","unstructured":"S. A. Siewior R. Russell S. Vaddagiri A. Raj J. Schopp and T. Gleixner. CPU hotplug in the Kernel \u2013 The Linux Kernel documentation. https:\/\/docs.kernel.org\/core-api\/cpu_hotplug.html"},{"key":"e_1_3_2_26_2","unstructured":"Performance boost at game loading time | Android Open Source Project. https:\/\/source.android.com\/docs\/core\/perf\/boost"},{"key":"e_1_3_2_27_2","doi-asserted-by":"crossref","unstructured":"E. Cortez A. Bonde A. Muzio M. Russinovich M. Fontoura and R. Bianchini. 2017. Resource central: Understanding and predicting workloads for improved resource management in large cloud platforms. In SOSP.","DOI":"10.1145\/3132747.3132772"},{"key":"e_1_3_2_28_2","doi-asserted-by":"crossref","unstructured":"Y. Zhang and Y. Huang. 2019. \u2018Learned\u2019: Operating systems. In ACM SIGOPS Operating Systems Review.","DOI":"10.1145\/3352020.3352027"},{"key":"e_1_3_2_29_2","doi-asserted-by":"crossref","unstructured":"R. Bitirgen E. Ipek and J. F. Mart\u00ednez. 2008. Coordinated management of multiple interacting resources in chip multiprocessors: A machine learning approach. In MICRO.","DOI":"10.1109\/MICRO.2008.4771801"},{"key":"e_1_3_2_30_2","volume-title":"FIRM: An intelligent fine-grained resource management framework for SLO-oriented microservices. In OSDI.","author":"Qiu H.","year":"2020","unstructured":"H. Qiu, S. S. Banerjee, S. Jha, Z. T. Kalbarczyk, and R. K. Iyer. 2020. FIRM: An intelligent fine-grained resource management framework for SLO-oriented microservices. In OSDI."},{"key":"e_1_3_2_31_2","unstructured":"S. Levy R. Yao Y. Wu Y. Dang P. Huang Z. Mu P. Zhao T. Ramani N. K. Govindaraju X. Li Q. Lin G. L. Shafriri and M. Chintalapati. 2020. Predictive and adaptive failure mitigation to avert production cloud VM interruptions. In OSDI."},{"key":"e_1_3_2_32_2","unstructured":"H.-D. Cho K. Chung and T. Kim. 2012. Benefits of the big.LITTLE Architecture. SAMSUNG Electronics White Paper."},{"key":"e_1_3_2_33_2","unstructured":"D. P. Kingma and J. Ba. Adam: A method for stochastic optimization. arXiv 1412.6980."},{"key":"e_1_3_2_34_2","unstructured":"Hyperopt documentation. http:\/\/hyperopt.github.io\/hyperopt\/"},{"key":"e_1_3_2_35_2","unstructured":"L. Liu X. Dou and Y. Chen. 2023. Intelligent resource scheduling for co-located latency-critical services: A multi-model collaborative learning approach. In FAST."},{"key":"e_1_3_2_36_2","unstructured":"TikTok: Vid\u00e9os LIVE Musique - Apps on Google Play. https:\/\/play.google.com\/store\/apps\/details?id=com.zhiliaoapp.musically&hl=en_US&pli=1"},{"key":"e_1_3_2_37_2","unstructured":"Android Debug Bridge (adb) | Android Studio | Android developers. https:\/\/developer.android.com\/tools\/adb"},{"key":"e_1_3_2_38_2","unstructured":"Weibo - Apps on Google Play. https:\/\/play.google.com\/store\/apps\/details?id=com.sina.weibo&hl=zh&gl=US"},{"key":"e_1_3_2_39_2","unstructured":"Taobao - Apps on Google Play. https:\/\/play.google.com\/store\/apps\/details?id=com.taobao.taobao&hl=en_US"},{"key":"e_1_3_2_40_2","unstructured":"TouTiao APP. https:\/\/app.toutiao.com\/"},{"key":"e_1_3_2_41_2","unstructured":"Jingdong - Apps on Google Play. https:\/\/play.google.com\/store\/search?q=jingdong&c=apps&fpr=false&gl=FR&hl=en_US"},{"key":"e_1_3_2_42_2","unstructured":"Watermelon video. https:\/\/sj.qq.com\/appdetail\/com.ss.android.article.video?fromcase=10003&from_wxz=1"},{"key":"e_1_3_2_43_2","unstructured":"ViPSHOP - Apps on Google Play. https:\/\/play.google.com\/store\/apps\/details?id=me.GooApp.Apps.ViPSHOP&hl=en_US&gl=FR"},{"key":"e_1_3_2_44_2","doi-asserted-by":"crossref","unstructured":"J. Haj-Yahya M. Alser J. S. Kim A. G. Yaglik\u00e7i N. Vijaykumar E. Rotem and O. Mutlu. 2020. SysScale: Exploiting multi-domain dynamic voltage and frequency scaling for energy efficient mobile processors. In ISCA.","DOI":"10.1109\/ISCA45697.2020.00029"},{"key":"e_1_3_2_45_2","doi-asserted-by":"crossref","unstructured":"S. A. L. Carvalho D. C. Cunha and A. G. Silva-Filho. 2019. Autonomous power management in mobile devices using dynamic frequency scaling and reinforcement learning for energy minimization. In Microprocessors and Microsystems.","DOI":"10.1016\/j.micpro.2018.09.008"},{"key":"e_1_3_2_46_2","doi-asserted-by":"crossref","unstructured":"S. Dey S. Saha A. K. Singh and K. D. McDonald-Maier. 2021. Asynchronous Hybrid Deep Learning (AHDL): A deep learning based resource mapping in DVFS enabled mobile MPSoCs. In WF-IoT.","DOI":"10.1109\/WF-IoT51360.2021.9594956"},{"key":"e_1_3_2_47_2","volume":"201","author":"Kim S.","unstructured":"S. Kim and Y. J. Kim. 2015. GPGPU-Perf: Efficient, interval-based DVFS algorithm for mobile GPGPU applications. In The Visual Computer.","journal-title":"J. Kim."},{"key":"e_1_3_2_48_2","doi-asserted-by":"crossref","unstructured":"Q. Wang and X. Chu. 2020. GPGPU performance estimation with core and memory frequency scaling. In IEEE TPDS.","DOI":"10.1145\/3366428.3380767"},{"key":"e_1_3_2_49_2","doi-asserted-by":"crossref","unstructured":"Y. Yao and Z. Lu. 2020. Pursuing extreme power efficiency with PPCC guided NoC DVFS. In IEEE TC.","DOI":"10.1109\/TC.2019.2949807"},{"key":"e_1_3_2_50_2","doi-asserted-by":"crossref","unstructured":"H. Yu Y. Ha B. Veeravalli F. Chen and H. El-Sayed. 2021. DVFS-based quality maximization for adaptive applications with diminishing return. In IEEE TC.","DOI":"10.1109\/TC.2020.2997242"},{"key":"e_1_3_2_51_2","doi-asserted-by":"crossref","unstructured":"J. Guerreiro A. Ilic N. Roma and P. Tomas. 2018. GPGPU power modeling for multi-domain voltage-frequency scaling. In HPCA.","DOI":"10.1109\/HPCA.2018.00072"},{"key":"e_1_3_2_52_2","unstructured":"OPPO browser review. https:\/\/communityin.oppo.com\/thread\/48606"},{"key":"e_1_3_2_53_2","unstructured":"Camera for OPPO. https:\/\/play.google.com\/store\/apps\/details?id=com.perffectselfie.thelastedcamera.oppocamerastyle&hl=en_US"},{"key":"e_1_3_2_54_2","unstructured":"Gmail - Apps on Google Play. https:\/\/play.google.com\/store\/apps\/details?id=com.google.android.gm&hl=en_US"},{"key":"e_1_3_2_55_2","unstructured":"Google Maps - Apps on Google Play. https:\/\/play.google.com\/store\/search?q=google%20maps&c=apps&hl=en_US"},{"key":"e_1_3_2_56_2","unstructured":"Amazon Shopping - Apps on Google Play. https:\/\/play.google.com\/store\/apps\/details?id=com.amazon.mShop.android.shopping&hl=en_US"},{"key":"e_1_3_2_57_2","unstructured":"WeChat - Apps on Google Play. https:\/\/play.google.com\/store\/search?q=wechat&c=apps&hl=en_US"},{"key":"e_1_3_2_58_2","unstructured":"The hall VR. https:\/\/cecropia.github.io\/thehallaframe\/"},{"key":"e_1_3_2_59_2","unstructured":"J. Chen A. Schranzhofer and L. Thiele. 2019. Energy minimization for periodic real-time tasks on heterogeneous processing units. In IPDPS."},{"key":"e_1_3_2_60_2","unstructured":"C. Yang J. Chen T. Kuo and L. Thiele. 2009. An approximation scheme for energy-efficient scheduling of real-time tasks in heterogeneous multiprocessor systems. In DATE."},{"key":"e_1_3_2_61_2","unstructured":"Lei Liu Yong Li Chen Ding Hao Yang and Chengyong Wu. 2006. Rethinking memory management in modern operating system: Horizontal vertical or random? In IEEE TC."},{"key":"e_1_3_2_62_2","doi-asserted-by":"crossref","unstructured":"L. Liu S. Yang L. Peng and X. Li. 2019. Hierarchical hybrid memory management in OS for tiered memory systems. In IEEE TPDS.","DOI":"10.1109\/TPDS.2019.2908175"}],"container-title":["ACM Transactions on Architecture and Code Optimization"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3674910","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3674910","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T00:05:56Z","timestamp":1750291556000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3674910"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,20]]},"references-count":61,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2025,3,31]]}},"alternative-id":["10.1145\/3674910"],"URL":"https:\/\/doi.org\/10.1145\/3674910","relation":{},"ISSN":["1544-3566","1544-3973"],"issn-type":[{"value":"1544-3566","type":"print"},{"value":"1544-3973","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,20]]},"assertion":[{"value":"2023-10-24","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2024-05-20","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-20","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}