The frontend web components were not the primary cause of high end user response times related to the loan approval process.
Business loan processing and personal loan processing need to be separated into multiple independent services so they can be independently scaled and maintained.
The need to have a more scalable, long-term archive storage solution for the loan audit data.
The collocation of loan data and audit data was constraining the database and needed to be split into independent repositories.
The need to have monitoring intelligence down to the infrastructure layer for each component accessible within the same UI.
AWS offers multiple services for container orchestration and storage. After careful consideration, the decision was made by the team to move forward with the following AWS stack that would best meet their technical and business requirements derived during the Mobilization phase.
Container Orchestration Selection
AWS Elastic Kubernetes Service (EKS)
Storage Solution Selection
AWS Simple Storage Service (S3)
Database Solution Selection
AWS Relational Database Service (RDS)
For container orchestration EKS aligns well with the teams requirements, having the ability to automate scaling, managing, updating, and removing containers at will without incurring any system downtime.
S3 (Simple Storage Solution) was chosen to host the audit data as it provides an affordable and robust solution to securely archive the needed audit trail while supporting high retrieval and ingestion rates with the ability to scale dynamically.
AWS RDS was chosen as the new database backend as it provides the high availability and fault tolerance needed with real time replication across multiple availability zones and/or regions.
Now it’s time to create the EKS cluster, deploy the application to the cluster, and look at how the AppDynamics agents are configured and deployed along with the application.