About the Company
The client provides contract lifecycle management software that legal and procurement teams use to draft, negotiate, and track agreements through their full lifecycle. The platform's databases hold the contracts and metadata the product is built around, which makes their cost, reliability, and performance central to the business.
The Challenge
The platform ran on commercially licensed Oracle databases the company hosted itself on Amazon EC2. That meant paying for Oracle licensing on top of running the engine day to day: patching, backups, and high availability all handled by the team. Capacity was fixed and sized by hand, so the company was left choosing between paying for more headroom than it usually needed and risking strain when activity spiked. The goals were straightforward: remove the licensing burden, take database operations off the team's plate, and move to a database that scales with demand without giving up performance.
The Solution
NileForge began by assessing the Oracle estate, the schema, stored procedures, and the dependencies between them, and used the AWS Schema Conversion Tool to convert the schema and PL/SQL into the PostgreSQL equivalents. The tool handled the bulk of the conversion automatically, and the team manually remediated the Oracle-specific constructs it could not translate directly, so nothing was left to chance.
For the data and the cutover, AWS Database Migration Service moved the data and then kept the new database continuously in sync with the live Oracle system through ongoing change-data-capture replication. That meant the switch over could happen in a short, planned window rather than a long outage, with the source available as a fallback until the new platform was confirmed healthy.
The databases were rebuilt on Amazon Aurora PostgreSQL and deployed on Amazon Aurora Serverless v2, which adjusts compute and memory automatically in fine-grained steps, so the platform absorbs peaks and scales back when things are quiet instead of paying continuously for maximum capacity. Aurora read replicas carry reporting and read-heavy traffic, and automated backups and Multi-AZ deployment cover resilience. The whole estate is defined as code in Terraform and maintained through CI/CD pipelines, replacing the manual setup with something consistent and repeatable. As an ongoing managed-services engagement, NileForge monitors performance with Amazon CloudWatch and Amazon RDS Performance Insights and keeps spend in view through AWS Cost Explorer and AWS Budgets, so the savings hold as the product grows.
The Results
- Commercial Oracle licensing eliminated by moving to Aurora PostgreSQL, removing a large recurring cost
- Capacity scales automatically with demand on Aurora Serverless v2, ending the overprovision-or-strain tradeoff
- Hands-on database maintenance replaced by a managed, automated, repeatable estate
- A near-zero-downtime cutover, with the live platform kept in sync until the planned switch
- Performance held under peak load, with reporting served by Aurora read replicas
Trade licensing for elasticity.
If commercial database licensing and hands-on maintenance are weighing on your platform, NileForge can migrate you to Amazon Aurora and keep your spend right-sized without giving up performance. Talk to our team
