TrustServing: A Quality Inspection Sampling Approach for Remote DNN Services
TrustServing: A Quality Inspection Sampling Approach for Remote DNN Services
复制标题
DOI：
10.1109/secon48991.2020.9158444
复制
发表时间：
2020-06
期刊：
影响因子：
--
通讯作者：
Xueyu Hou;T. Han
中科院分区：
文献类型：
--
作者：
Xueyu Hou;T. Han
深度神经网络（DNN）正被应用于计算机视觉、自动驾驶汽车和医疗保健等各个领域，然而，DNN因其高计算复杂度而臭名昭著，并且无法在资源受限的物联网（IoT）设备上有效执行。已经提出了各种解决方案来处理DNN的高计算复杂度。将DNN的计算任务从物联网设备卸载到云/边缘服务器是最受欢迎和最有前途的解决方案之一。虽然由服务器提供的这种远程DNN服务大大减少了IoT设备上的计算任务，但IoT设备检查服务质量是否满足其服务水平目标（SLO）是具有挑战性的。在本文中，我们解决了这个问题，并提出了一种名为QIS（质量检查抽样）的新方法，可以有效地检查物联网设备的远程DNN服务的质量。为了实现QIS，我们设计了一种新的ID生成方法来生成可以识别边缘服务器上的DNN模型的数据（ID）。QIS将标识插入到输入数据流中，并对SLO违规进行抽样检查。实验结果表明，QIS方法可以可靠地检测，接近100%的成功率，当SLA水平为99.9%或更低的远程DNN服务的服务质量，成本只有0.5%。
Deep neural networks (DNNs) are being applied to various areas such as computer vision, autonomous vehicles, and healthcare, etc. However, DNNs are notorious for their high computational complexity and cannot be executed efficiently on resource constrained Internet of Things (IoT) devices. Various solutions have been proposed to handle the high computational complexity of DNNs. Offloading computing tasks of DNNs from IoT devices to cloud/edge servers is one of the most popular and promising solutions. While such remote DNN services provided by servers largely reduce computing tasks on IoT devices, it is challenging for IoT devices to inspect whether the quality of the service meets their service level objectives (SLO) or not. In this paper, we address this problem and propose a novel approach named QIS (quality inspection sampling) that can efficiently inspect the quality of the remote DNN services for IoT devices. To realize QIS, we design a new ID-generation method to generate data (IDs) that can identify the serving DNN models on edge servers. QIS inserts the IDs into the input data stream and implements sampling inspection on SLO violations. The experiment results show that the QIS approach can reliably inspect, with a nearly 100% success rate, the service qualtiy of remote DNN services when the SLA level is 99.9% or lower at the cost of only up to 0.5% overhead.