Architecting on AWS Certification Equipment Maintenance & Calibration 5 — Questions and Answers
Question 1: An architect wants to calibrate CloudFront cache behavior so that dynamic API responses are never cached while static assets are cached for 24 hours. How should cache policies be configured?
- Set a single default TTL of 0 for all origins
- Use separate cache behaviors with CachingDisabled policy for the API path and a custom TTL policy for static paths (Correct answer)
- Use Lambda@Edge to set Cache-Control headers
- Enable Origin Shield for all origins
Correct answer: Use separate cache behaviors with CachingDisabled policy for the API path and a custom TTL policy for static paths
CloudFront supports multiple cache behaviors per distribution; assigning CachingDisabled to the /api/* path and a custom TTL policy to /static/* provides precise per-path cache control.
Question 2: A company runs a fleet of IoT-connected industrial sensors that send telemetry to AWS IoT Core. The team needs to calibrate alert thresholds centrally and push updated configuration to all devices. Which feature enables this?
- AWS IoT Device Shadow
- AWS IoT Jobs (Correct answer)
- AWS IoT Greengrass component deployment
- Amazon Kinesis Data Streams
Correct answer: AWS IoT Jobs
AWS IoT Jobs allows you to define and send remote operations — such as configuration updates or firmware pushes — to a set of devices, tracking execution status per device.
Question 3: An operations team must ensure that AWS Lambda function concurrency is capped at 200 to prevent downstream database overload during traffic spikes. How should this be configured?
- Set account-level reserved concurrency to 200
- Configure a reserved concurrency limit of 200 on the specific Lambda function (Correct answer)
- Use Lambda Provisioned Concurrency set to 200
- Apply an SQS queue in front of the function with a batch size of 200
Correct answer: Configure a reserved concurrency limit of 200 on the specific Lambda function
Setting a reserved concurrency limit on a specific Lambda function ensures it never scales beyond that threshold, protecting downstream resources from overload.
Question 4: A data engineering team needs to recalibrate Kinesis Data Streams throughput by adjusting the shard count in response to sustained backlog growth. What is the correct approach?
- Delete and recreate the stream with more shards
- Use the UpdateShardCount API to rescale the stream (Correct answer)
- Increase the retention period to absorb the backlog
- Switch to Kinesis Data Firehose
Correct answer: Use the UpdateShardCount API to rescale the stream
The UpdateShardCount API allows online resharding of a Kinesis Data Stream, increasing or decreasing shard count without data loss or stream recreation.
Question 5: During a maintenance window, an architect needs to temporarily suspend an Auto Scaling group from launching or terminating instances while manual work is performed. Which action achieves this?
- Set desired capacity to 0
- Suspend the Launch and Terminate scaling processes (Correct answer)
- Detach all instances from the group
- Disable all CloudWatch alarms
Correct answer: Suspend the Launch and Terminate scaling processes
Suspending the Launch and Terminate processes in an Auto Scaling group halts all scale-out and scale-in activities while preserving the existing running instances.
Question 6: A team must verify that their multi-region active-active architecture fails over correctly before a production maintenance event. Which AWS service provides pre-built runbooks for testing and executing failover procedures?
- AWS Fault Injection Service (FIS)
- AWS Systems Manager Automation with AWSSupport runbooks
- AWS Resilience Hub
- Amazon Route 53 Application Recovery Controller (Correct answer)
Correct answer: Amazon Route 53 Application Recovery Controller
Amazon Route 53 Application Recovery Controller provides readiness checks, routing controls, and safety rules specifically designed to safely execute and validate multi-region failover.
Question 7: An architect must ensure that Elastic Beanstalk environments receive platform updates (OS patches, runtime upgrades) during a defined maintenance window with minimal downtime. Which update type should be selected?
- Immutable update policy
- Rolling with additional batch update policy
- Managed platform updates with instance replacement (Correct answer)
- All-at-once update policy
Correct answer: Managed platform updates with instance replacement
Elastic Beanstalk Managed Platform Updates apply patch-level and minor updates automatically during a configured maintenance window, replacing instances in a rolling fashion to minimize downtime.
Architecting on AWS Certification Equipment Maintenance & Calibration 5
An architect wants to calibrate CloudFront cache behavior so that dynamic API responses are never cached while static assets are cached for 24 hours.
How should cache policies be configured?