移动云计算环境下协同资源迁移机制研究

Nowadays, mobile devices such as smartphones and tablets are quickly becoming the prominent computing and communication platform, which allows billions of people to interact with each other and access to diverse resources and information through wireless networks efficiently and ubiquitously. The convergence of mobile computing and cloud computing forms the Mobile Cloud Computing (MCC) paradigm, which provides a uniform platform for resource-limited mobile devices to access a pool of virtualized resources, and enables a range of new computation-intensive and content-centric applications such as healthcare, online video gaming, and crowd sensing. This project focuses on the key mechanisms of cooperative resource offloading for mobile cloud computing in order to reduce energy consumption and system overhead. First, we propose efficient dynamic context measurement and application partition methods, which forms the context information model and data flow graph model for mobile applications. Then based on the derived models, we study the key techniques of resource offloading, which includes multi-level cooperation for resource sharing and multi-objective optimization for resource offloading decision. Finally, we combine the above techniques to form a comprehensive resource offloading system for mobile cloud and evaluate its performance using smartphones. The result of our research can be applied to a wide range of scenarios such as mobile device augmentation, distributed content distribution, massive storage, mobile big data processing and analysis.

近年来，移动设备如智能手机和平板电脑等已迅速成为流行的计算和通信平台，是人们日常交互和随时随地获取信息服务的主要工具。移动计算和云计算技术的融合形成了新的移动云计算架构，它为资源受限的移动设备提供了统一的平台来访问虚拟化的资源池，从而拓展了移动设备的能力，可以支持计算和存储密集的新型应用如健康监护、在线视频游戏、群智感知等。本项目研究移动云计算环境下高效协同的资源迁移机制。首先，提出动态情境信息测量和应用任务划分的方法，构建了移动应用运行的情境信息模型和数据流图模型。其次，基于上述模型，研究云资源迁移的关键技术，包括多层面的协同资源共享机制和多目标优化的资源迁移决策算法。最后，综合各方面优化结果，形成高效的移动云资源迁移体系并在智能手机上进行性能评估。本项目的研究成果在移动设备增强，分布式内容分发、海量存储、移动大数据处理和分析等方面具有广泛应用前景。

近年来，移动设备如智能手机和平板电脑等已迅速成为流行的计算和通信平台，是人们日常交互和随时随地获取信息服务的主要工具。移动计算和云计算技术的融合形成了新的移动云计算架构，它为资源受限的移动设备提供了统一的平台来访问虚拟化的资源池，从而拓展了移动设备的能力以支持计算和存储密集的新型应用如健康监护、在线视频游戏、群智感知等。本项目研究面向移动云计算的高效协同资源迁移机制。首先，提出动态情境信息测量和应用任务划分的方法，构建了移动应用运行的情境信息模型和数据流图模型。其次，基于上述模型，研究云资源迁移的关键技术，包括多层面的协同资源共享机制和多目标优化的资源迁移决策算法。最后，综合各方面优化结果，形成高效的移动云资源迁移体系并在智能手机上进行性能评估。..以本项目工作为基础，课题组成员在包括IEEE Journal on Selected Areas in Communications (JSAC)，IEEE Transactions on Communications， IEEE Transactions on Network and Service Management (TNSM)，Computer Networks，Journal of Computer Science and Technology (JCST)，INFOCOM 2017，INFOCOM 2019，IMWUT/Ubicomp 2018，IMWUT/Ubicomp 2019，ACM Multimedia 2019，ACM Multimedia 2020，IJCAI 2018，IJCAI 2019等国际期刊和国际会议录用和发表论文 28篇，其中 SCI 检索论文6篇，CCF-A类论文10篇，CCF-B类论文7篇。提交国家专利申请 10项。发表英文专著2章节。在该项目资助下，共培养博士生5名、硕士生20名。.