MetaCity: An Edge Emulator with the Feature of Realistic Geospatial Support for Urban Computing.

The edge computing paradigm is an important supplement to the traditional cloud computing paradigm in current IoT application scenarios. However, edge computing is highly related to a specific application scenario, in which the mobility of edge devices and the geographical distribution of edge infrastructure are strongly correlated. However, it is expensive to deploy the solution in the real world and most current edge computing emulators lack realistic scenario support and mobility support. Therefore, it is challenging to evaluate whether an edge infrastructure deployment solution can satisfy the QoS (Quality-of-Service) requirement of an edge application in a cost-effective manner. In this paper, we propose and implement an edge emulator, MetaCity, which is able to effectively enforce edge computing policies and construct realistic application scenarios. MetaCity can leverage geographical data to establish an emulation environment according to the realistic infrastructure deployment strategy and emulate the mobility process of edge devices based on the actual urban road network. MetaCity can also provide an extensible network QoS monitoring module that supports the concurrent execution of various QoS monitoring in an emulated environment. In addition, MetaCity provides a user-friendly web-based graphical user interface instead of text-based configuration files. For evaluation, three smart transportation-based experiments are conducted to validate the functionality, scalability, and emulation accuracy of MetaCity.