Date of Award
Master of Science (MS)
Through this dissertation we have explored the concept that is Edge Computing and have endeavored to outline the design of a framework around which edge computing applications can be built. All edge computing applications being distributed in nature, the framework itself has to be distributed at least between two devices - one being the user or client side and the other the server side. With multiple clients in the picture, a client can take up the role of a secondary "server" in the traditional sense and this makes the intended platform a truly collaborative one. The primary server hosts a service or client registry and keeps a track of connected devices through heart beat messages. The processing centers on the server are miniature destructible machines called Docker containers and the design of the processing unit follows a scalable Factory design pattern. The framework on the client side consists of interfaces to the several units on the server. There is one for service registry and discovery, one for communicating with the processing unit and a third one for the heart beats. There is one last interface on the other side of the client to talk to Bluetooth enabled devices. We have evaluated the framework by developing two applications and measuring the latency of data transfer. While the work put behind the dissertation was fruitful, it goes without saying that this is a work in progress and much work needs to be done specially to address security and privacy concerns in collaborative applications that the framework intends to cater to.
Padhee, Subir Kumar, "Edge Computing Framework: A Platform for Smart Phone Users, IoT Peripherals and Edge Servers to Collaborate" (2017). Electrical, Computer & Energy Engineering Graduate Theses & Dissertations. 152.