CSR: Small: Extending Smartphone Battery Life via Prescriptive Energy Profiling

CSR：小：通过规范的能量分析延长智能手机电池寿命

• 批准号: 1718854
• 负责人: Charlie Hu
• 金额: $ 47.5万
• 依托单位: Purdue University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-07-01 至 2020-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1718854&HistoricalAwards=false
• 关键词: CSR; Small; Extending; Smartphone; Battery

With over 2 billion users worldwide and over 2 million applications (apps) each in Google and Apple's digital marketplaces, smartphones have advanced human society into the mobile era. Despite the far-reaching societal impact of smartphones, the user experience has been severely limited by smartphone battery life. The obvious idea of extending smartphone battery life is to improve the battery technology, which, however, has stagnated in the past 15 years, and battery capacity is generally viewed as nearing the limits of physics. Instead, this project pursues a course of research to significantly extend smartphone battery life by reducing the energy drain of various software running on the smartphone.This project aims to tackle a major smartphone battery drain contributor, app energy drain when running in the foreground, with the goal of reducing it by 2x. The proposed research includes major thrusts that will develop the first set of prescriptive energy profiling techniques that will transform energy profiling from a "passive" app development step into a "proactive" process: (a) It will develop the first capability to automatically uncover the causal relationship between parameter configurations and framework/library method invocations, a unique and hard program analysis problem in framework-based mobile programming; (b) It will develop the first capability to automatically identify an energy-efficient implementation of similar tasks, a generally intractable problem but attainable for mobile apps by comparing similar apps that are widely available in the huge mobile app marketplace.The importance of this project is heightened by smartphones being an important enabler of Internet access for disadvantaged people in both developed and developing countries, and hence being an important tool in overcoming the "digital divide". Technically, this work will have a far-reaching impact outside of the area of developing energy efficient mobile apps by increasing compiler capabilities for general event-driven computer programs. A project repository will be maintained to provide access to code, raw experimental data showcasing the effectiveness of the developed techniques, and processed results and technical publications. The repository will be maintained during the three-year project period and the two years afterwards, to foster research activities by others in this important area.

智能手机在全球拥有超过 20 亿用户，Google 和 Apple 数字市场中各有超过 200 万个应用程序 (app)，推动人类社会进入移动时代。 尽管智能手机具有深远的社会影响，但用户体验却受到智能手机电池寿命的严重限制。延长智能手机电池寿命的明显想法是改进电池技术，然而，电池技术在过去15年里一直停滞不前，电池容量普遍被认为接近物理极限。相反，该项目致力于通过减少智能手机上运行的各种软件的能量消耗来显着延长智能手机电池寿命的研究过程。该项目旨在解决智能手机电池消耗的一个主要贡献者，即应用程序在前台运行时的能量消耗，减少 2 倍的目标。拟议的研究包括开发第一套规范性能源分析技术的主要推动力，该技术将能源分析从“被动”应用程序开发步骤转变为“主动”过程：（a）它将开发第一个自动发现能源分析的能力。参数配置与框架/库方法调用之间的因果关系，是基于框架的移动编程中独特且困难的程序分析问题； (b) 它将开发第一个自动识别类似任务的节能实施的能力，这通常是一个棘手的问题，但通过比较庞大的移动应用市场中广泛使用的类似应用程序，可以实现移动应用程序。该项目的重要性智能手机成为发达国家和发展中国家弱势群体接入互联网的重要推动者，因此成为克服“数字鸿沟”的重要工具。 从技术上讲，这项工作将通过提高通用事件驱动计算机程序的编译器功能，在开发节能移动应用程序领域之外产生深远的影响。将维护一个项目存储库，以提供对代码、展示所开发技术有效性的原始实验数据以及处理结果和技术出版物的访问。该存储库将在三年项目期间和之后的两年内得到维护，以促进其他人在这一重要领域的研究活动。

项目成果

Differential Energy Profiling: Energy Optimization via Diffing Similar Apps

差分能量分析：通过比较类似应用程序进行能量优化

• 发表时间: 2018-10
• 期刊: 13th USENIX Symposium on Operating Systems Design and Implementation
• 作者: Jindal, Abhilash;Hu, Y. Charlie
• 通讯作者: Hu, Y. Charlie