Efficient Distributed Threshold-Based Offloading for Large-Scale Mobile Cloud Computing
Efficient Distributed Threshold-Based Offloading for Large-Scale Mobile Cloud Computing
复制标题
DOI：
10.1109/tnet.2022.3193073
复制
发表时间：
2023-02
期刊：
影响因子：
--
通讯作者：
Xudong Qin;Bin Li;Lei Ying
中科院分区：
文献类型：
--
作者：
Xudong Qin;Bin Li;Lei Ying
移动云计算使计算受限的移动设备能够通过利用强大的云服务器来执行实时密集型计算，例如语音识别或对象检测。大规模移动云计算中的一个重要问题是计算卸载，每个移动设备都会通过考虑本地处理延迟以及使用云服务器的成本来决定何时以及将多少计算上传到云服务器。在本文中，我们开发了一个基于分布式阈值的卸载算法，如果在设备上排队的任务数量达到阈值并在本地处理，则它将传入的计算任务上传到云服务器。阈值根据计算负载和使用云服务器的成本进行迭代更新。我们将问题提出为对称游戏，并表征了假设指数服务时间的NASH平衡（NE）的存在和独特性的足够和必要条件。然后，我们显示了我们提出的分布式算法与NE存在时的收敛性。此外，当使用云服务器的成本很高时，我们表征了提议的分布式算法下的成本与无政府状态价格（POA）的最低成本之间的成本之间的性能差距。最后，我们执行广泛的模拟来验证我们的理论发现，证明我们在各种情况下提出的分布式算法的效率，例如过度的服务时间，服务器使用不完美的服务估计以及异步阈值更新，并揭示基于阈值的质量质量的优越性能。
Mobile cloud computing enables compute-limited mobile devices to perform real-time intensive computations such as speech recognition or object detection by leveraging powerful cloud servers. An important problem in large-scale mobile cloud computing is computational offloading, where each mobile device decides when and how much computation should be uploaded to cloud servers by considering the local processing delay and the cost of using cloud servers. In this paper, we develop a distributed threshold-based offloading algorithm where it uploads an incoming computing task to cloud servers if the number of tasks queued at the device reaches the threshold and processes it locally otherwise. The threshold is updated iteratively based on the computational load and the cost of using cloud servers. We formulate the problem as a symmetric game, and characterize the sufficient and necessary conditions for the existence and uniqueness of the Nash Equilibrium (NE) assuming exponential service times. Then, we show the convergence of our proposed distributed algorithm to the NE when the NE exists. Further, we characterize the performance gap between cost under our proposed distributed algorithm and the minimum cost in terms of Price of Anarchy (PoA) when the cost of using cloud servers is high. Finally, we perform extensive simulations to validate our theoretical findings, demonstrate the efficiency of our proposed distributed algorithm under various scenarios such as hyperexponential service times, imperfect server utilization estimation, and asynchronous threshold updates, and reveal the superior performance of threshold-based policies over their probabilistic counterpart.