One of the most important requirements for an edge/cloud computing system is task latency, especially for some specific applications such as security alarm services and actuator feedback control in the Industrial Internet of Things (IIoT).

In this paper, the authors present a method to minimize service delay, which results in two physical cloudlet servers. Different from traditional methods that focus on either communication or computation delay, this paper exploits a way to improve both elements with an integration technique. Several mathematical models are further presented to evaluate the system delay as the sum of transmission delay and processing delay. Analysis results show the proposed “dual focus approach is the best way to tackle the service delay problem in edge cloud computing.”