A Layer-Based Runtime Framework for Enabling Cloud Computing into Fog Stratum

Abstract

Evolution of computing paradigms has been one of the main reasons of IT revolution over the past few decades. Cloud computing has been playing a major role in this revolution, offering different kinds of demands as services, such as Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS). Cloud computing helps in solving many issues of datacenters as resource utilization, resource sharing and scalability. Yet, the cloud computing still has some challenges tackling latency-sensitive applications such as Fire detection, Firefighting, Healthcare applications, robot control and content delivery. This is caused mainly because of the connectivity between the cloud and the end devices. Another reason can be the intra-communication delay occurring between different cloud instances to provide the final service or answer. Fog Computing is a new computing paradigm that has been introduced to offer solutions for the cloud computation challenges. Introducing a new conceptual layer between Cloud computing and Edge devices called Fog opened new horizons for computation offloading close to the consumers. Offloading computations near the Edge layer plays an important role in reducing the latency of the services. There is a lack of Fog application development tools and frameworks that are easily adopted by software engineers in the software community. In this thesis, we provide tools and frameworks that utilizes the Fog stratum in an easily abstracted manner for software engineers. Two main contributions are proposed in our thesis to tackle the lack of fog stratum utilization. A Fog based CDN framework is proposed to facilitate using fog nodes as a CDN nodes. Also, a new annotation and meta language is proposed along with its compiler to facilitate generating a service that can be deployed in fog architecture. These tools and framework provide better quality of service (QoS) for cloud services in terms of latency and reliability.