Core Banking Applications Refactoring and Cloud Migration

Problem

The bank operated multiple on‑premise monolithic core‑banking applications. Six of these critical apps were assigned to the project. Their monolithic nature led to long, costly deployment cycles that limited the bank’s ability to respond quickly to market demands. Frameworks and libraries also required updates to safeguard application security.

Project Objectives

• Move to a cloud‑native architecture with zero downtime (24/7 service continuity).
• Reduce infrastructure Total Cost of Ownership (TCO) by at least 20 %.
• Accelerate release cycles via a CI/CD pipeline.
• Consolidate authentication through the bank’s corporate Single Sign‑On (SSO).
• Improve application security by updating frameworks and libraries.

Solution Applied

• Refactored the applications to run efficiently in the cloud, integrating persistent volumes with S3‑compatible storage, optimizing communications, and performing database and batch‑service migrations or adjustments using PHP Vanilla, Yii, Symfony and Laravel.
• Migrated and optimized MySQL, PostgreSQL and Oracle databases for optimal performance and efficient data management.
• Containerized workloads with Docker and Podman and deployed them on Kubernetes, facilitating management and scalability.
• Integrated the bank’s corporate SSO across all applications, unifying authentication.
• Implemented a CI/CD pipeline with blue‑green deployments and automated tests to accelerate the release cadence.

Technologies Applied

PHP Vanilla · Yii · Symfony · Laravel · Docker · Podman · Kubernetes · OpenAPI · S3‑compatible Storage · MySQL · PostgreSQL · Oracle

Results & Benefits

• 23 % infrastructure TCO reduction in the first year, surpassing the target.
• 40 % faster deployments, moving from quarterly to weekly releases thanks to CI/CD.
• Almost zero downtime during migration via blue‑green strategy, maintaining 24/7 service.
• Enhanced security and user experience through SSO and framework updates.
• 30 % lower database query latency, improving transaction performance.