A financial services company is using SAP Event Mesh to broadcast real-time trade confirmations. A downstream auditing application needs to consume every message exactly once, even if the application is temporarily offline. A separate market data application needs to receive all messages but can tolerate occasional loss. Which configuration should be used?
Answer and explanation
Correct answer: C
This is the correct publish-subscribe pattern. Publishing to a topic allows multiple, independent consumers. The auditing application subscribes via a queue to ensure message persistence and guaranteed, exactly-once delivery (with QoS 1). The market data application subscribes directly to the topic with QoS 0 (at-most-once delivery), which is suitable for its non-critical requirement.
Question 2
During a security audit of an API Proxy in SAP API Management, a consultant observes that a 'Verify API Key' policy is attached to the ProxyEndpoint Preflow. However, a requirement states that a specific resource path, /public/status, should be accessible without an API key, while all other paths must be secured. How can this be achieved without creating a separate API Proxy?
Answer and explanation
Correct answer: D
Policies in the Preflow apply to all requests. To create an exception, the security policy must be moved out of the Preflow. By placing the 'Verify API Key' policy in the default flow (which acts as a catch-all) and creating a specific conditional flow for /public/status with no security policies, the requirement is met. The specific flow for /public/status is matched and executed, and processing continues without hitting the default flow's security policy.
Question 3
A developer is building an integration flow that processes a batch of sales orders from an SFTP server. The source file is a large XML containing thousands of elements. The flow must process each order individually, call an external OData service to enrich it with customer data, and then send it to a target system. To optimize performance and memory usage, which combination of iFlow steps is most appropriate?
Answer and explanation
Correct answer: B
The Iterating Splitter is designed for large messages, as it processes splits sequentially without loading the entire message into memory, preventing memory-related errors. The Content Enricher is the correct step to call an external service to add data to the current message (the individual order). The Gather step is used to combine the enriched messages back into a single message if required, or can be omitted if orders are sent individually after enrichment.
Question 4
When defining an integration strategy for a company adopting a 'Clean Core' approach for their SAP S/4HANA Cloud system, which of the following activities is considered a direct violation of this principle?
Answer and explanation
Correct answer: D
The core principle of a 'Clean Core' is to avoid modifications to the standard SAP code. Custom logic should be implemented as side-by-side extensions on SAP BTP or as in-app extensions using public, released APIs. Directly modifying standard ABAP programs creates dependencies that complicate upgrades and maintenance, which is precisely what the Clean Core strategy aims to prevent.
Question 5
Multiple answers
A company wants to expose a single, unified API for its customers that aggregates data from three different backend microservices. They need to enforce a single API key for authentication, apply a consistent rate limit, and transform the three backend responses into one consolidated JSON payload. Which two SAP Integration Suite capabilities are best suited for this task? (Select TWO)
Answer and explanation
Correct answers: A, C
Question 6
True or False: In SAP Event Mesh, a single queue can be configured to subscribe to multiple different topics.
Answer and explanation
Correct answer: A
This is a key feature of SAP Event Mesh for building flexible event-driven architectures. A single queue can have multiple topic subscriptions, allowing it to collect events from various sources (topics) into one consolidated stream for a consumer application. This is useful for aggregating related events, for example, 'OrderCreated', 'OrderUpdated', and 'OrderCancelled' events into a single 'OrderEvents' queue.
Question 7
Case Study
A global logistics company, ShipFast, is modernizing its integration landscape by moving from a legacy SAP PI/PO system to SAP Integration Suite. A critical interface involves processing large EDI 850 Purchase Order files from their partners, which are delivered via SFTP. These files can contain up to 10,000 individual line items.
The current business process requires that each incoming EDI 850 file is first validated against a schema. If validation fails, the entire file must be moved to an error directory and a high-priority alert sent to the support team. If validation succeeds, each line item must be processed individually. For each line item, the integration flow must look up the product's current stock level from an on-premise S/4HANA system via an OData API. If stock is insufficient for a particular line item, that specific item must be logged to a data store for later review, but processing of other items in the same file should continue. All successfully validated and enriched line items must be aggregated and sent as a single XML message to an internal order management system via an HTTPS call.
The key non-functional requirements are high throughput, minimal memory footprint on the Cloud Integration runtime, and detailed error logging for failed line items. The S/4HANA system has a strict rate limit on its OData API, allowing only 100 requests per minute.
Which integration flow design best meets all the requirements of ShipFast?
Answer and explanation
Correct answer: B
This design correctly addresses all requirements. The XML Validator handles the initial file-level validation. The ProcessDirect call modularizes the design. The Iterating Splitter is essential for processing large files with low memory usage. Placing the OData call (Content Enricher) inside a loop with a delay respects the rate limit (100 requests/min ≈ 1 request/600ms). The Router correctly separates successful items from those with insufficient stock. The Data Store Write step fulfills the requirement to log failed items for later review without stopping the main process. Finally, the Gather step correctly aggregates the successful items for the final HTTPS call.
Question 8
An integration developer needs to store a value, such as a transaction counter, that must persist and be accessible across multiple, independent executions of the same integration flow. Which iFlow component should be used for this purpose?
Answer and explanation
Correct answer: C
Data Stores are specifically designed for stateful processing. They allow data to be persisted beyond the lifetime of a single message processing run. Variables and properties are message-scoped and are lost once the integration flow execution is complete. A Data Store allows the flow to write a value (like a counter) and a subsequent, independent execution of the same flow to read and update that value.
Question 9
Multiple answers
An architect is designing a solution where events from an SAP S/4HANA system must trigger workflows in multiple, independent subscribing applications. Some subscribers might be temporarily unavailable and must not lose any events. The solution must ensure that the S/4HANA system is not tightly coupled to the subscribers. Which components should be used to build this solution? (Select THREE)
Answer and explanation
Correct answers: A, C, D
Question 10
A developer needs to apply a sequence of three policies to an API Proxy in SAP API Management: 1. Quota, 2. Spike Arrest, 3. Verify API Key. To ensure the most efficient processing and to avoid unnecessary policy execution for unauthorized requests, what is the optimal order and location for attaching these policies?
Answer and explanation
Correct answer: B
The optimal order is to fail as early as possible. Authentication/authorization (Verify API Key) should happen first to immediately reject invalid requests. Traffic management policies should follow. Spike Arrest should come before Quota, as it deals with immediate traffic bursts and can protect the system from DoS attacks, while Quota deals with usage over a longer period (e.g., per day or month). Checking the API key first prevents unauthorized requests from consuming any Spike Arrest or Quota counts.
