Refactoring Monolithic Application into Cloud-Native Architecture

Abstract

Cloud Native is an approach, using which, applications are developed and run in such a way as to exploit and use the maximum features of cloud computing. Going cloud native does not just mean deploying your code to the cloud but following some rules and patterns to develop the application from the start. We want to evaluate the performance of an on-premise application with a parallel cloud native application through experimentation. We developed a cloud native application from scratch, using Platform as a service (PaaS) by Google cloud platform (GCP), following all the patterns that will be described in this report. GCP provides PaaS to develop,run and manage your applications without having to manage the infrastructure. This cloud native application is a counterpart of a monolithic and on-premises application called DUP (Delivery Platform). The performance analysis is based on metrics e.g. zero-downtime deployment, continuous deployment, automation of DevOps, extensibility of microservices, effective provisioning and efficient roll out strategies. We discover that the cloud native architecture performs better under these metrics.