Provisioning of Disaster Recovery with Terraform and Kubernetes: A Case Study on Software Defect Prediction
Rizal Broer Bahaweres, Farhan Muhammad Najib · 2023
Natural disasters and device failures can damage organizational infrastructure and data, giving substantial issues for modern firms that rely on information systems and technology, especially cloud-based solutions. To maintain firm operations, information and technology systems must be protected. The conventional catastrophic recovery technique trades speed, complexity, and cost. Infrastructure as a Code (IaC) may improve complex disaster recovery operations in firms. This study analyzes Infrastructure as Code as a potential solution to cloud disaster recovery issues. The study examines how Terraform and Kubernetes affect cloud-based disaster recovery procedures and durations by simulating a disaster event. The findings of the study indicate that the implementation of Terraform for constructing intricate infrastructure, such as a deployment pipeline, has a notable impact on the reduction of required steps and the time required. The number of steps is significantly reduced to a single step and reduce the time required by 235% when compared to manual manufacturing. Terraform demonstrates its ability to optimise disaster recovery. The utilization of Kubernetes has been found to significantly enhance the efficiency of software deployment. The use of Kubernetes resulting in a notable reduction in deployment time by approximately 530%. This improvement in deployment speed is expected to have implications for the software recovery process in the event of a disaster. The duration of catastrophe recovery using Terraform and Kubernetes is 190% shorter compared to the duration of manual recovery. It reduces the use of unutilized surplus infrastructure resources and makes rebuilding easier, but it takes longer than standby infrastructure.