SAP-C02 Practice Question 20: Message Broker Replatform
AWS Certified Solutions Architect – Professional (SAP-C02) · Test 2 · question 10 of 10
A logistics company is migrating a portfolio of Java applications to AWS. About 40 of them exchange messages through a self-managed Apache ActiveMQ Classic broker on premises, using JMS over the OpenWire protocol with queues, topics, durable subscriptions, and request/reply patterns. The broker is a single server, and patching it has caused several outages.
The migration plan classifies the messaging layer as replatform. Applications must keep their JMS code and client libraries unchanged apart from connection configuration. The broker must become a managed service that AWS patches during a defined maintenance window. The layer must survive the loss of an Availability Zone with automatic failover, and messages must be stored durably across multiple Availability Zones.
Which target should the solutions architect choose?
Show answer and explanation
Answer:
Amazon MQ is a managed message broker for Apache ActiveMQ Classic and RabbitMQ. AWS positions it for migrating existing brokers without rewriting messaging code: the same industry-standard protocols and JMS APIs keep working, and Amazon MQ applies maintenance and version upgrades during the broker's maintenance window. That is a textbook replatform. An active/standby deployment runs two brokers in two different Availability Zones as a redundant pair backed by Amazon EFS, which stores data redundantly across multiple Availability Zones. If the active broker fails, the standby becomes active. Amazon MQ provides a pair of wire-protocol endpoints, and clients connect through the ActiveMQ failover transport so they reconnect to whichever endpoint is active. Only the connection URL changes, which the stem allows.
Option B is a refactor, not a replatform. The SQS Java Messaging Library implements only a subset of JMS, point-to-point queues, so topics and durable subscriptions would have to be redesigned around SNS. OpenWire clients would also need their libraries replaced.
Option C is a rehost that fails the managed-service requirement. The team would still patch ActiveMQ and the operating system itself and design its own failover. Putting a load balancer in front of an active/standby ActiveMQ pair is also not how its failover works.
Option D is managed and uses durable multi-AZ storage, but a single-instance broker is one broker in one Availability Zone. If that zone fails, there is no broker to serve clients until it recovers, so the automatic-failover requirement is not met. Durable storage is not availability.
References: What is Amazon MQ?; Deployment options for Amazon MQ for ActiveMQ brokers