Cost-effective Offloading Strategy for Delay-sensitive Applications in Cloud of Things Systems

The steep rise of Internet of Things (IoT) applications along with the limitations of Cloud Computing to address all IoT requirements leveraged a new distributed computing paradigm called Fog Computing, which aims to process data at the edge of the network. With the help of Fog Computing, some of the uncertainties of the communication among different tiers like the transmission latency, monetary spending and application loss caused by Cloud Computing can be effectively reduced. However, as the processing capacity of Fog nodes is more limited than that of cloud platforms, running all applications indiscriminately on these nodes can cause some QoS requirement to be violated. Therefore, we apply Three-tier Cloud of Things Systems (including Things Tier, Fog Tier and Cloud Tier), a simple yet powerful model of IoT to do the applications offloading between Fog Tier and Cloud Tier to guarantee some processing requirements. And there is important decision-making as to where executing each application in order to produce a cost effective solution and fully meet application requirements. This thesis makes the original contribution of an appropriate dynamical optimization about the application offloading. We call this the study of online decision making and this study is based on queueing theory. Additionally, we clarify the relationship between processing time and QoS requirements (like the money cost) to provide cost-effective application processing service. In particular, we are interested in the tradeoff in terms of average response time, average cost and average number of application loss. In this thesis, we present an online algorithm, called Unit-slot Optimization, based on the technique of Lyapunov Optimization. The Unit-slot Optimization is a quantified near-optimal online solution to balance the two-way tradeoff between average response time and average cost. Meanwhile, the proposed optimization is satisfied with the three-way tradeoff among average response time, average cost and average number of application loss as well. After that, we evaluate the performance of the Unit-slot Optimization algorithm by a number of experiments. The experimental results not only match up the theoretical analyses properly, but also demonstrate that our proposed algorithm can successfully provide cost-effective processing, while guaranteeing average response time in a Three-tier Cloud of Things System