Assuring 2000 payment transactions per second reliably with resiliency testing

Large financial institutions lose money and customers when transactions drop, especially during rapid growth in customer base, fast developing teams, and technological innovation. Chaos engineering has emerged as a trusted testing discipline that changed the game for large companies as it involves testing not only the deployments but also the application infrastructure to see how systems will respond to real-world failures. This would mean improving reliability and resiliency, reducing financial loss and increasing customer satisfaction.

Our client, a large FinTech company in Asia processing over 1 billion payment transactions a day, faced a similar issue where financial transactions dropped during peak demands on the system. This resulted in financial loss, a decrease in customer satisfaction, and a toil on developers responding to incidents. They needed a solution that improved the reliability and resiliency of the overall platform.

Nagarro proposed a chaos engineering solution to improve the resilience of the client's FinTech platform. The team built a Continuous Resilience Testing platform that automates chaos experiments in delivery pipelines and production, leveraging the shift-right approach to run the tests in production. The company's chaos engineering efforts resulted in a significant improvement in the reliability and resiliency of its platform. This led to a reduction in financial loss, an increase in customer satisfaction, and a decrease in the toil on developers.