The recent increase in the number of Internet of Things (IoT) devices has led to the generation of a large amount of data. These data are generally processed by cloud servers because of their high scalability and ability to provide resources on demand. However, processing large amounts of data in the cloud is an impractical solution for the strict requirements of IoT services, such as low latency and high bandwidth. Fog computing, which brings computational resources closer to the IoT devices, has emerged as a suitable solution to mitigate these problems. Resource provisioning and application orchestration are two of the key challenges when running IoT applications in a Fog computing environment. In this paper, we present ElasticFog, which runs on top of the Kubernetes platform and enables real-time elastic resource provisioning for containerized applications in Fog computing. Specifically, ElasticFog collects network traffic information in real time and allocates computational resources proportionally to the distribution of network traffic. The experimental results prove that ElasticFog achieves a significant improvement in terms of throughput and network latency compared with the default mechanism in Kubernetes.

Container-based virtualization is becoming a de facto way to build and deploy applications because of its simplicity and convenience. Kubernetes is a well-known open-source project that provides an orchestration platform for containerized applications. An application in Kubernetes can contain multiple replicas to achieve high scalability and availability. Stateless applications have no requirement for persistent storage; however, stateful applications require persistent storage for each replica. Therefore, stateful applications usually require a strong consistency of data among replicas. To achieve this, the application often relies on a leader, which is responsible for maintaining consistency and coordinating tasks among replicas. This leads to a problem that the leader often has heavy loads due to its inherent design. In a Kubernetes cluster, having the leaders of multiple applications concentrated in a specific node may become a bottleneck within the system. In this paper, we propose a leader election algorithm that overcomes the bottleneck problem by evenly distributing the leaders throughout nodes in the cluster. We also conduct experiments to prove the correctness and effectiveness of our leader election algorithm compared with a default algorithm in Kubernetes.