Free AWS AIP-C01 Actual Exam Questions - Question 15 Discussion

A/C? A sounds neat with multi-variant endpoints handling traffic splits internally, cutting down manual work. But C’s blue/green with ALB is a proven pattern for smooth rollouts and real-time traffic control, plus monitoring can hook into ALB easily. B feels more complex with multiple endpoints and API Gateway management, which can add overhead. D seems the least operationally efficient since manual ALB weight changes don’t scale well. Between A and C, it’s about whether you prefer built-in SageMaker routing or the flexibility (and familiarity) of blue/green deployments with ALB. Both minimize

Maybe A, since having all models behind one endpoint simplifies traffic routing a lot, and you avoid juggling multiple services. The built-in traffic weighting feels more seamless than manually tweaking ALB or API Gateway rules.

It’s B because using separate endpoints with API Gateway lets you plug in custom monitoring easily and adjust traffic weights without worrying about SageMaker’s built-in limits. More flexible for real-time metrics and automation.

Option B avoids the complexity of multi-variant endpoints and lets you leverage API Gateway’s weighted routing, which can be easier to integrate with custom monitoring tools for real-time metric tracking and traffic shifting.

A/C? A is great for easy traffic splits, but C’s blue/green deployment with ALB might offer smoother 100% traffic shifts based on metrics. Still, C probably needs more setup though.

It’s A for me too. Using SageMaker multi-variant endpoints means you don’t have to juggle multiple endpoints or build a custom routing layer like with B or D. Also, the traffic weighting is built-in, so less manual work. While real-time engagement monitoring and auto-shifting might need some outside setup, that’s still way less overhead than managing blue/green deployments and ALBs like in C. Overall, A strikes the best balance of simplicity and efficiency for this use case.

It’s A. The key is operational efficiency with minimal overhead, and SageMaker multi-variant endpoints let you manage all model versions behind one endpoint with simple traffic weight adjustments. While it might not handle real-time engagement metrics fully on its own, integrating with CloudWatch or other monitoring tools can automate shifts without complex infrastructure. Options B and D definitely require more manual routing, and C adds deployment complexity with CodeDeploy and ALB. So A fits best for seamless traffic routing and easier management in production.

I don’t think B or D fit since they both seem to require more manual traffic management. A uses built-in traffic weights which makes it more operationally efficient than C. So I’d go with A.

Option A seems best since it reduces overhead by managing all models behind one endpoint.

A also seems solid because AWS DMS is designed for database replication and supports secure connections. It specifically mentions replicating only non-sensitive data, which matches the requirement to keep sensitive info in-house. Plus, DMS can work over an IPsec connection if set up with a VPN, so it fits the secure transfer demand. B and D feel off since B relies on Lambda filtering after streaming, which risks sensitive data exposure, and D transfers the full backup, which violates the sensitive data rule. So, A looks like a good fit alongside C.

C (Glue filters, VPN ensures IPsec security, fits requirements best).

C looks best since AWS Glue can filter non-sensitive data and Site-to-Site VPN covers the IPsec secure transfer requirement. A and B don’t clearly confirm IPsec, and D transfers the whole backup, which includes sensitive data.