Every successful SaaS platform eventually faces the same “good problem to have”: your application is so popular that its own growth becomes the biggest threat to its stability. This is the story of how our team at Jurnal (Mekari) tackled a major performance bottleneck in our core Ruby on Rails monolith, and how we evolved our architecture iteratively to support our rapidly growing user base.
When I was at Jurnal, our core product was a feature-rich accounting platform built on a classic, robust stack: a Ruby on Rails monolith backed by a MySQL database on AWS RDS, with Redis and Sidekiq for caching and background job processing. All traffic came through an AWS Load Balancer and an API Gateway (Kong) to manage our growing third-party ecosystem.
