Task 2.3: Design and Implement Enterprise Integration Architectures. AIP-C01 Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   A large insurance company uses a mainframe-based policy administration system that only supports COBOL copybook data formats over MQ Series. The company wants to add a generative AI feature that summarizes policy documents for customer service agents. The AI service is exposed as a REST API that accepts and returns JSON. The integration team must minimize changes to the mainframe while enabling the AI service to be called from existing mainframe workflows. What should the team do?

   • ADeploy an API gateway in front of the AI service and configure the mainframe to call the gateway directly using REST.
   • BUse AWS Database Migration Service to replicate policy data to Amazon S3 and have the AI service read directly from S3.
   • CCreate a middleware service that translates MQ messages to REST calls and converts COBOL copybook data to JSON, then expose it to the mainframe as an MQ queue.
   • DReplace the mainframe's MQ interface with a modern REST client and update all policy data to JSON format.
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   A legacy enterprise resource planning (ERP) system uses SOAP/XML for all external communication. The team wants to integrate it with a foundation model that exposes a REST/JSON API. What is the primary integration challenge to address?

   • AMigrating the ERP system to a modern cloud-native architecture before integration.
   • BData format conversion between XML and JSON, and protocol adaptation from SOAP to REST.
   • CEnsuring the foundation model has enough GPU capacity to handle the ERP's request volume.
   • DReplacing the foundation model's REST API with a SOAP API to match the ERP.
   Reveal Solution Discussion Select an answer first
4. 18application · medium

   A healthcare organization maintains a patient records database in Amazon Aurora. A generative AI application generates clinical summaries that must be stored back into the same database for downstream reporting. The summaries are produced asynchronously by a worker process. The team wants to ensure that the summaries are written to Aurora without overloading the database and that the data remains consistent with the original patient records. Which pattern should the team use?

   • AUse AWS Database Migration Service to continuously replicate Aurora data to Amazon S3 and have the worker write summaries to S3.
   • BHave the worker process write each summary directly to Aurora using a new database connection per request.
   • CUse Amazon Kinesis Data Streams to stream summaries directly into Aurora via an AWS Lambda function that writes each record.
   • DPublish a completion event to an Amazon SQS queue and have a batch consumer that periodically writes accumulated summaries to Aurora using a connection pool.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   A financial services firm wants to integrate a foundation model that generates investment insights into its existing customer relationship management (CRM) system. The CRM is a legacy application that only supports SOAP-based web services. The integration must be synchronous because the CRM user interface waits for the insight to display. The team needs to expose the FM capability to the CRM in a way that minimizes changes to the CRM. What should the team do?

   • AModify the CRM to call the FM API directly using REST and JSON.
   • BCreate a SOAP-to-REST proxy service that translates SOAP requests to REST calls to the FM API and returns the response in SOAP format.
   • CExport CRM data to Amazon S3 and run a nightly batch job that generates insights and imports them back.
   • DUse an Amazon SQS queue to asynchronously send requests to the FM and poll for the response.
   Reveal Solution Discussion Select an answer first
6. 20application · medium

   A retail bank has a legacy core banking system that uses IBM MQ for messaging. The bank wants to integrate a foundation model that can detect fraudulent transactions in real time. The FM service is exposed as a REST API. The integration must be synchronous because the fraud detection result is needed before the transaction is approved. What should the team do?

   • AUse AWS Lambda to call the FM API, but have the core banking system invoke Lambda synchronously via the AWS SDK.
   • BModify the core banking system to call the FM API directly using REST.
   • CCreate a middleware service that listens on an MQ queue, translates the message to a REST call, and sends the response back to another MQ queue.
   • DUse an Amazon SQS queue to asynchronously send the transaction to the FM and poll for the result.
   Reveal Solution Discussion Select an answer first