An Agent-Based Fog Computing Architecture for Resilience on Amazon EC2 Spot Instances

Cloud computing providers have started offering their idle resources as transient servers. Spot instances are transient servers offered by Amazon, whose prices dynamically change over time based on supply and demand. By using appropriate strategies and fault-tolerant mechanisms, users can effectively use spot instances to run applications at a lower price. This paper presents a resilient agent-based fog computing architecture that combines machine learning and a statistical model to predict time to instance revocation and helps to refine fault tolerance parameters and reduce total execution time. The experiments demonstrate that our model predicts with high levels of accuracy reaching 94% success rate what indicates the model is effective under realistic working conditions.

I. Introduction

In recent years the Cloud Computing paradigm has evolved to offer secure services to execute applications in an environment with high scalability and flexibility and affordable pricing for users. Thus, cloud computing has emerged as an important community resource and is considered a new alternative for the execution of applications with high levels of security and reliability [1]. Cloud services and resources are available in a managed and virtual platform that dynamically provides scalable resources in a transparent way to users. A particular class of service is Infrastructure as a Service (IaaS), which offers a set of Virtual Machines (VMs) of different types and capacity, allowing users to choose according their needs.