SAP Integration Developer practical mock exam: solution and scoring rubric

A complete practice assessment with a fictional scenario — an API Management task via Cloud Connector and an autonomous Cloud Integration task with context changes and a Data Store. Includes timing, steps and a 100-point rubric.

After passing the System-based Assessment, I created this practice assessment with two independent tasks, mandatory artifact names and checklist validation. The scenario is fictional, preserving the confidentiality of the real exam. Solving it under time pressure is a useful preparation exercise.

How to use it: reserve 90 uninterrupted minutes, open your Integration Suite trial tenant and avoid reading the solution until you finish or decide to stop. Score your work with the rubric at the end.

Scenario

Aurora Foods is a food distributor with an on-premise ERP. It needs two deliverables from you as an Integration Developer:

• Task A: expose the ERP's material catalogue OData service to business partners through API Management, using an already configured Cloud Connector connection.
• Task B: build an internal Cloud Integration process that flattens a warehouse hierarchy into a simple SKU list and saves it in internal storage.

Required names are exact and case-sensitive: AUR_Proxy ≠ aur_proxy. Write them down before starting.

Task A — Expose the on-premise catalogue with API Management

Requirement

Publish the ERP's ZMAT_CATALOG_SRV OData service in the Developer Hub so partners can subscribe using an API Key.

Technical details

Required artifacts: exact names

What is assessed

Understand the chain and its order: the Provider defines the technical connection, using On Premise and Location ID to route through Cloud Connector. The Proxy exposes the service discovered from the Provider's catalogue. The Product packages the Proxy and is published to the Developer Hub. Creation alone is insufficient: the Proxy must be Deployed and the Product Published.

Task B — Internal flattening pipeline with Cloud Integration

Requirement

Build an autonomous iFlow that runs when deployed, without a sender or external call. It takes hierarchical inventory — warehouse → aisle → SKU — transforms it into a flat SKU list, ignores intermediate levels and persists the result in a Data Store before ending.

Input payload: place it at the start of the flow

<?xml version="1.0" encoding="UTF-8"?>
<inv:WarehouseInventory xmlns:inv="http://aurora-foods.example/inventory">
  <Warehouse Name="Central">
    <Aisle Number="A1">
      <SKU>ARZ-1001</SKU>
      <SKU>ARZ-1002</SKU>
    </Aisle>
    <Aisle Number="A2">
      <SKU>ACE-2001</SKU>
      <SKU>ACE-2002</SKU>
      <SKU>ACE-2003</SKU>
    </Aisle>
  </Warehouse>
  <Warehouse Name="Norte">
    <Aisle Number="N1">
      <SKU>HAR-3001</SKU>
      <SKU>HAR-3002</SKU>
    </Aisle>
    <Aisle Number="N2">
      <SKU>AZU-4001</SKU>
    </Aisle>
  </Warehouse>
</inv:WarehouseInventory>

Expected Data Store result

A single-level list containing eight SKUs:

<inv:SKUList xmlns:inv="http://aurora-foods.example/inventory">
  <SKU>ARZ-1001</SKU>
  <SKU>ARZ-1002</SKU>
  <SKU>ACE-2001</SKU>
  <SKU>ACE-2002</SKU>
  <SKU>ACE-2003</SKU>
  <SKU>HAR-3001</SKU>
  <SKU>HAR-3002</SKU>
  <SKU>AZU-4001</SKU>
</inv:SKUList>

Required artifacts: exact names

Create the mapping XSD/WSDL from the structures above, or simulate using editor types. The assessed concept is context handling rather than schema typing.

⏱️ Stop reading and solve the tasks. The solution follows below.

Task A solution

1. Configure → APIs → API Providers → Create: name AUR_Provider. Under Connection, choose On Premise, host erp.aurora-foods.internal, port 44300, and Location ID plant001. This value routes through the corresponding Cloud Connector. Under Catalog Service Settings, configure the path prefix, catalogue URL and Basic authentication from the table.
2. Create API → API Provider source → AUR_Provider → Discover → select ZMAT_CATALOG_SRV → name AUR_Proxy → Deploy. Creating a proxy from a pasted direct URL does not establish the Provider→Proxy relationship required by this exercise.
3. Engage → Products → Create: AUR_Product → APIs tab → Add → AUR_Proxy → Publish. Without publication, partners cannot find and subscribe to the product in the Developer Hub.

In a trial without a real ERP, Test Connection will fail. You can still practice the configuration sequence. For a working end-to-end lifecycle using a simulated CPI backend, follow the governed API case.

Task B solution

1. Design → Create package AUR_Package → Add → Integration Flow AUR_Flow → Edit.
2. Autonomous means Timer: remove the Sender participant, Start Message and channel. Add a Timer start event with Run Once scheduling so it runs upon deployment.
3. Add a Content Modifier after the Timer. Under Message Body, select Constant and paste the complete XML payload. Copying it avoids namespace typing mistakes.
4. Add Message Mapping, creating the artifact as SKUFlattener. Keep the exact required name. Source structure: WarehouseInventory; target: SKUList.
5. The key step — change the context: map Warehouse/Aisle/SKU → SKUList/SKU. An initial simulation may group SKUs by aisle or show a context error. Select source node SKU in the expression editor and change its context to root WarehouseInventory, or use node function removeContexts. This removes intermediate Warehouse/Aisle contexts so all SKUs enter one list. Simulate and verify eight flat SKUs.
6. Add Data Store Operations → Write after mapping: Data Store Name AUR_Store, empty Entry ID for automatic generation, Integration Flow visibility. Connect it to End, ensuring the write happens before completion.
7. Save → Deploy. Verify a Completed message in Monitor → Message Processing, then one entry in Manage Stores → Data Stores → AUR_Store containing the flat list.

Self-assessment rubric: 100 points

80+ points: strong performance on this practice rubric. 60–79: review failed criteria and repeat in three days. Below 60: revisit the fundamentals using the laboratory cases. This rubric is a practice aid, not an official exam score or guarantee.

Frequent mistakes and how to avoid them

1. Using your own artifact names instead of the required names: validation is literal.
2. Leaving the Proxy undeployed or Product unpublished: creation alone does not finish the task.
3. Leaving a Sender in the iFlow: this autonomous exercise requires a Timer.
4. Mapping without changing context: grouped output can look correct at first glance. Simulate and count elements.
5. Skipping final verification in Data Store and Monitor: finish by checking the actual result.

Share your score on LinkedIn or contact me. If you are preparing for certification, read my exam experience and explore the free resources. 🎯