How do you approach database partitioning and scaling for transactional data that grows exponentially at Adyen?
