Question

A financial services company is building an order-processing platform composed of independent microservices (Payment, Inventory, Shipping) deployed to Azure Container Apps. Each service owns its own database. A single business operation spans all three services, and if any step fails, prior steps must be logically undone to maintain data consistency. The company wants to avoid distributed locks and long-held transactions across services. Which architectural pattern should the solution architect recommend to coordinate the operation and handle failures?

Accepted Answer

Implement the Saga pattern using a sequence of local transactions where each service publishes an event, and compensating transactions reverse completed steps on failure

Answer

Implement a two-phase commit (2PC) coordinator that holds a distributed transaction across all three service databases until every participant commits

Answer

Consolidate the three databases into a single shared database so all operations run inside one ACID transaction

Answer

Use the Cache-Aside pattern with Azure Cache for Redis to store operation state and roll back by clearing the cache on failure