Exam AZ-305 topic 4 question 29 discussion

The given answer is correct. ADF pipeline can process the message and trigger the appropriate condition. On ADF, you can add a diagnostic setting to send logs to a storage account. Other possible options would be Event grid subscription & Service bus topic. Service bus TOPIC can be used with filtering rules on each subscription but not queue. https://learn.microsoft.com/en-us/azure/service-bus-messaging/service-bus-queues-topics-subscriptions#rules-and-actions

Option A looks correct to me: an Azure Service Bus queue ADF pipeline is for data ETL/movement only

Azure Service Bus queue = best choice. Factory pipeline is for data movement and transformation not for event-driven integration.

Repost from @d717098: I would like to meet the az-305 certified architect who recommends Azure Data Factory as an integration component in a message based architecture, honestly... ADF for Web App decoupling, seriously?

Service Bus - messaging service used for Order processing and financial transactions. ref: https://learn.microsoft.com/en-us/azure/service-bus-messaging/compare-messaging-services

ADF is not for messaging. Azure Event Grid domain is not ideal when message ordering, delivery guarantees, or control logic are required. Also, Event Grid is push-based, not queue-based. Azure Event Hubs capture is used for event streaming and telemetry ingestion, not for transactional message processing or logic routing.

The best choice for the integration component in this order processing system would be Azure Service Bus. Why Azure Service Bus? Reliable Asynchronous Messaging – Ensures that order requests are processed even if vendors are temporarily unavailable. Message Routing & Workflows – Allows publish-subscribe messaging, where App1 sends requests to topics, and vendors subscribe to relevant orders. Triggering Azure Functions – Service Bus can trigger Function1 or Function2 when an order is processed. Decoupled Communication – Vendors and functions do not need to be tightly coupled to App1, improving flexibility. Guaranteed Message Delivery – Ensures that messages are persisted, even in cases of failures or temporary disconnections. Alternative: Azure Event Grid (If Event-Driven Architecture is Preferred) Best for real-time event notifications, rather than queued messaging. Suitable if vendors need instant updates without queuing orders.

ADF is not used for queue processing but voor ETL/ELT loading. So Service bus makes more sence.

Azure Service Bus queue

Regardless of how easily ADF pipelines can handle a single use case, from arch perspective, it cannot be a solution. ETL / data transformation pipelines are designed for batch processing with large amounts of data rather than high numbers of messages. A successful ordering app could easily exceed its concurrency max capacity... On the other hand, it will be much pricier. On the other hand, this is a classical use case for event driven architectures, service bus is designed for these use cases.

Service Bus is a fully managed enterprise message broker with message queues and publish-subscribe topics. The service is intended for enterprise applications that require transactions, ordering, duplicate detection, and instantaneous consistency. https://learn.microsoft.com/en-us/azure/service-bus-messaging/compare-messaging-services

Data Factory is an ETL/ELT tool, not a messaging system. Hence, “an Azure Service Bus queue” is the most appropriate. - o1

Service Bus Queue makes the most sense given the scenario.

"Message processing and triggering functions (Function1 or Function2)" Event-driven workflow → Service Bus queues are built for message-based integrations. Ensures reliable message delivery, including retries, ordering, and dead-lettering. "Checking product availability at vendor 1 and vendor 2" Requires request-response messaging, best handled by Azure Service Bus (queue or topic). Service Bus ensures that each message is processed once and reliably. "Transaction steps are logged to storage" Service Bus supports transactional processing, ensuring all steps are logged correctly.

Service Bus Queue is build to hand this use case. It is a message handling service. You need to have a guarantee and low latency for order processing systems. Data Factory use case is ETL.

I go for service bus queue, since ADF is typically used for batch processing and scheduled workflows, not for handling individual messages or event-driven architectures.

The key bit for me on this one is “An integration component will process the message, and then trigger either Function1 or Function2 depending on the type of order.”. We’re being asked to name the integration component. Since service bus will only queue the data and not trigger functions, it has to be Azure Data Factory.