Automatic Repaving of Unhealthy Nodes in Cloud Infrastructure

Abstract

Cloud computing is a modern paradigm in which elastic services are provided to the users over the Internet. Scalability, availability and optimal resource utilization are the key aspects of cloud computing. In order to ensure availability of nodes, periodic health check must be done, and self-healing ability must be provided for each node to ensure customer doesn’t faces any downtime. Kubernetes is one Orchestration system that provides a structure for efficient operation of distributed containers. It takes care of auto-scaling, container failures, automates deployment patterns, & much more. Using Kubernetes API’s special module Repaver is written which provides Node healing and has been detailed in this paper.