Scaling Beyond a Billion Transactions Per Day with Sub-Second Responses

Abstract: Mastercard’s Decision Management Platform serves as the core for multiple business solutions to combat fraud and validate cardholder identity. The patented Java-based platform has grown over the years to process over a billion transactions a day while adhering to strict SLAs. To make accurate decisions, the platform keeps historical and real-time aggregates for billions of parameters. In this session, we’ll go through the journey of our transition to distributed data processing using GemFire and the challenges faced along the way. Technical topics include using server-side functions to enable scalable real-time compute, pros and cons of delta propagation, and avoiding update slowdowns due to replication latency.