Amazon AWS Certified Solutions Architect Associate SAA-C03 Queues Events and Workflow Decoupling Practice Test

 

Topic 07 covers queues events and workflow decoupling for the AWS Certified Solutions Architect – Associate certification. These original practice questions apply the verified SAA-C03 objectives to practical decisions and troubleshooting. Select one answer unless a fixed number is requested. For broader preparation, visit the AWS Certified Solutions Architect Associate SAA-C03 Exam Dumps page. Each option includes an explanation of the relevant behavior and scenario constraints.

Question 1

A web storefront emits one order task that must be processed by exactly one available worker, and producers must not wait for workers to be online. Which service is the best fit?

1. Use Amazon MQ.
2. Use Amazon EventBridge.
3. Use Amazon SQS.
4. Use Amazon SNS.
5. Use Amazon API Gateway.

Correct Answer: C

 

Correct Answer

Answer C is correct because SQS provides durable queueing that decouples producers from consumers and supports work-queue processing. This directly satisfies the decisive requirement in the scenario: durable decoupling.

Incorrect Answers

Answer A is incorrect because Amazon MQ provides managed message brokers compatible with established broker protocols such as ActiveMQ or RabbitMQ. It can be useful in other designs, but it does not satisfy the decisive requirement here: durable decoupling.

Answer B is incorrect because EventBridge routes events from producers to targets by rules on an event bus. It can be useful in other designs, but it does not satisfy the decisive requirement here: durable decoupling.

Answer D is incorrect because SNS provides publish/subscribe fan-out to multiple subscribers. It can be useful in other designs, but it does not satisfy the decisive requirement here: durable decoupling.

Answer E is incorrect because API Gateway provides managed HTTP, REST, and WebSocket API front doors. It can be useful in other designs, but it does not satisfy the decisive requirement here: durable decoupling.

 

Question 2

A monitoring platform must publish the same alarm notification to billing, security, and operations subscribers independently. Which service best matches the broadcast requirement?

1. Use Amazon API Gateway.
2. Use Amazon SQS.
3. Use Amazon SNS.
4. Use Amazon MQ.
5. Use Amazon EventBridge.

Correct Answer: C

 

Correct Answer

Answer C is correct because SNS provides publish/subscribe fan-out to multiple subscribers. This directly satisfies the decisive requirement in the scenario: pub/sub fan-out.

Incorrect Answers

Answer A is incorrect because API Gateway provides managed HTTP, REST, and WebSocket API front doors. It can be useful in other designs, but it does not satisfy the decisive requirement here: pub/sub fan-out.

Answer B is incorrect because SQS provides durable queueing that decouples producers from consumers and supports work-queue processing. It can be useful in other designs, but it does not satisfy the decisive requirement here: pub/sub fan-out.

Answer D is incorrect because Amazon MQ provides managed message brokers compatible with established broker protocols such as ActiveMQ or RabbitMQ. It can be useful in other designs, but it does not satisfy the decisive requirement here: pub/sub fan-out.

Answer E is incorrect because EventBridge routes events from producers to targets by rules on an event bus. It can be useful in other designs, but it does not satisfy the decisive requirement here: pub/sub fan-out.

 

Question 3

A company wants events from many AWS services and custom applications routed to different targets based on event content without coupling producers to subscribers. Which service should be used?

1. Use Amazon SQS.
2. Use Amazon EventBridge.
3. Use Amazon SNS.
4. Use Amazon API Gateway.
5. Use Amazon MQ.

Correct Answer: B

 

Correct Answer

Answer B is correct because EventBridge routes events from producers to targets by rules on an event bus. This directly satisfies the decisive requirement in the scenario: event bus.

Incorrect Answers

Answer A is incorrect because SQS provides durable queueing that decouples producers from consumers and supports work-queue processing. It can be useful in other designs, but it does not satisfy the decisive requirement here: event bus.

Answer C is incorrect because SNS provides publish/subscribe fan-out to multiple subscribers. It can be useful in other designs, but it does not satisfy the decisive requirement here: event bus.

Answer D is incorrect because API Gateway provides managed HTTP, REST, and WebSocket API front doors. It can be useful in other designs, but it does not satisfy the decisive requirement here: event bus.

Answer E is incorrect because Amazon MQ provides managed message brokers compatible with established broker protocols such as ActiveMQ or RabbitMQ. It can be useful in other designs, but it does not satisfy the decisive requirement here: event bus.

 

Question 4

A migrated application already speaks JMS and requires managed ActiveMQ-compatible broker semantics with minimal code changes. Which service is most appropriate?

1. Use Amazon API Gateway.
2. Use Amazon SQS.
3. Use Amazon MQ.
4. Use Amazon SNS.
5. Use Amazon EventBridge.

Correct Answer: C

 

Correct Answer

Answer C is correct because Amazon MQ provides managed message brokers compatible with established broker protocols such as ActiveMQ or RabbitMQ. This directly satisfies the decisive requirement in the scenario: managed message broker.

Incorrect Answers

Answer A is incorrect because API Gateway provides managed HTTP, REST, and WebSocket API front doors. It can be useful in other designs, but it does not satisfy the decisive requirement here: managed message broker.

Answer B is incorrect because SQS provides durable queueing that decouples producers from consumers and supports work-queue processing. It can be useful in other designs, but it does not satisfy the decisive requirement here: managed message broker.

Answer D is incorrect because SNS provides publish/subscribe fan-out to multiple subscribers. It can be useful in other designs, but it does not satisfy the decisive requirement here: managed message broker.

Answer E is incorrect because EventBridge routes events from producers to targets by rules on an event bus. It can be useful in other designs, but it does not satisfy the decisive requirement here: managed message broker.

 

Question 5

A mobile client needs a managed HTTPS endpoint for synchronous request/response calls to a backend service. Which service best provides the API front door?

1. Use Amazon SQS.
2. Use Amazon EventBridge.
3. Use Amazon API Gateway.
4. Use Amazon SNS.
5. Use Amazon MQ.

Correct Answer: C

 

Correct Answer

Answer C is correct because API Gateway provides managed HTTP, REST, and WebSocket API front doors. This directly satisfies the decisive requirement in the scenario: synchronous API invocation.

Incorrect Answers

Answer A is incorrect because SQS provides durable queueing that decouples producers from consumers and supports work-queue processing. It can be useful in other designs, but it does not satisfy the decisive requirement here: synchronous API invocation.

Answer B is incorrect because EventBridge routes events from producers to targets by rules on an event bus. It can be useful in other designs, but it does not satisfy the decisive requirement here: synchronous API invocation.

Answer D is incorrect because SNS provides publish/subscribe fan-out to multiple subscribers. It can be useful in other designs, but it does not satisfy the decisive requirement here: synchronous API invocation.

Answer E is incorrect because Amazon MQ provides managed message brokers compatible with established broker protocols such as ActiveMQ or RabbitMQ. It can be useful in other designs, but it does not satisfy the decisive requirement here: synchronous API invocation.

 

Question 6

A payment consumer sometimes receives the same SQS message more than once. The business operation must never charge twice. Which design change is most important?

1. Use an SQS FIFO queue with a meaningful MessageGroupId.
2. Set or extend the SQS visibility timeout to cover normal processing and extend it when needed.
3. Set the SQS message retention period to the required processing window.
4. Make the consumer operation idempotent by recording or deriving a unique business request identifier.
5. Configure a dead-letter queue with an appropriate redrive policy.

Correct Answer: D

 

Correct Answer

Answer D is correct because Idempotent processing prevents a redelivered message from causing a duplicate business side effect. This directly satisfies the decisive requirement in the scenario: duplicate-safe processing.

Incorrect Answers

Answer A is incorrect because FIFO queues serialize messages within each message group while allowing other groups to progress. It can be useful in other designs, but it does not satisfy the decisive requirement here: duplicate-safe processing.

Answer B is incorrect because The visibility timeout keeps an in-flight message hidden from other consumers while it is being processed. It can be useful in other designs, but it does not satisfy the decisive requirement here: duplicate-safe processing.

Answer C is incorrect because The retention period controls how long unconsumed messages remain available before SQS deletes them. It can be useful in other designs, but it does not satisfy the decisive requirement here: duplicate-safe processing.

Answer E is incorrect because A DLQ captures messages that repeatedly fail so they do not circulate indefinitely in the source queue. It can be useful in other designs, but it does not satisfy the decisive requirement here: duplicate-safe processing.

 

Question 7

Events for each customer must remain ordered, but different customers may be processed in parallel. Which SQS feature best supports this?

1. Configure a dead-letter queue with an appropriate redrive policy.
2. Use an SQS FIFO queue with a meaningful MessageGroupId.
3. Set or extend the SQS visibility timeout to cover normal processing and extend it when needed.
4. Set the SQS message retention period to the required processing window.
5. Make the consumer operation idempotent by recording or deriving a unique business request identifier.

Correct Answer: B

 

Correct Answer

Answer B is correct because FIFO queues serialize messages within each message group while allowing other groups to progress. This directly satisfies the decisive requirement in the scenario: message group identifiers.

Incorrect Answers

Answer A is incorrect because A DLQ captures messages that repeatedly fail so they do not circulate indefinitely in the source queue. It can be useful in other designs, but it does not satisfy the decisive requirement here: message group identifiers.

Answer C is incorrect because The visibility timeout keeps an in-flight message hidden from other consumers while it is being processed. It can be useful in other designs, but it does not satisfy the decisive requirement here: message group identifiers.

Answer D is incorrect because The retention period controls how long unconsumed messages remain available before SQS deletes them. It can be useful in other designs, but it does not satisfy the decisive requirement here: message group identifiers.

Answer E is incorrect because Idempotent processing prevents a redelivered message from causing a duplicate business side effect. It can be useful in other designs, but it does not satisfy the decisive requirement here: message group identifiers.

 

Question 8

A worker normally needs 8 minutes to process an SQS message, but the queue visibility timeout is 2 minutes and duplicate work appears. What should be changed?

1. Configure a dead-letter queue with an appropriate redrive policy.
2. Make the consumer operation idempotent by recording or deriving a unique business request identifier.
3. Use an SQS FIFO queue with a meaningful MessageGroupId.
4. Set or extend the SQS visibility timeout to cover normal processing and extend it when needed.
5. Set the SQS message retention period to the required processing window.

Correct Answer: D

 

Correct Answer

Answer D is correct because The visibility timeout keeps an in-flight message hidden from other consumers while it is being processed. This directly satisfies the decisive requirement in the scenario: extend visibility while processing.

Incorrect Answers

Answer A is incorrect because A DLQ captures messages that repeatedly fail so they do not circulate indefinitely in the source queue. It can be useful in other designs, but it does not satisfy the decisive requirement here: extend visibility while processing.

Answer B is incorrect because Idempotent processing prevents a redelivered message from causing a duplicate business side effect. It can be useful in other designs, but it does not satisfy the decisive requirement here: extend visibility while processing.

Answer C is incorrect because FIFO queues serialize messages within each message group while allowing other groups to progress. It can be useful in other designs, but it does not satisfy the decisive requirement here: extend visibility while processing.

Answer E is incorrect because The retention period controls how long unconsumed messages remain available before SQS deletes them. It can be useful in other designs, but it does not satisfy the decisive requirement here: extend visibility while processing.

 

Question 9

One malformed message fails every attempt and is blocking useful investigation. Which configuration should isolate it after repeated receives?

1. Configure a dead-letter queue with an appropriate redrive policy.
2. Use an SQS FIFO queue with a meaningful MessageGroupId.
3. Set the SQS message retention period to the required processing window.
4. Set or extend the SQS visibility timeout to cover normal processing and extend it when needed.
5. Make the consumer operation idempotent by recording or deriving a unique business request identifier.

Correct Answer: A

 

Correct Answer

Answer A is correct because A DLQ captures messages that repeatedly fail so they do not circulate indefinitely in the source queue. This directly satisfies the decisive requirement in the scenario: dead-letter queue.

Incorrect Answers

Answer B is incorrect because FIFO queues serialize messages within each message group while allowing other groups to progress. It can be useful in other designs, but it does not satisfy the decisive requirement here: dead-letter queue.

Answer C is incorrect because The retention period controls how long unconsumed messages remain available before SQS deletes them. It can be useful in other designs, but it does not satisfy the decisive requirement here: dead-letter queue.

Answer D is incorrect because The visibility timeout keeps an in-flight message hidden from other consumers while it is being processed. It can be useful in other designs, but it does not satisfy the decisive requirement here: dead-letter queue.

Answer E is incorrect because Idempotent processing prevents a redelivered message from causing a duplicate business side effect. It can be useful in other designs, but it does not satisfy the decisive requirement here: dead-letter queue.

 

Question 10

A compliance feed can be processed for only four days; messages older than that have no business value. Which queue setting should enforce this lifecycle?

1. Set or extend the SQS visibility timeout to cover normal processing and extend it when needed.
2. Configure a dead-letter queue with an appropriate redrive policy.
3. Set the SQS message retention period to the required processing window.
4. Make the consumer operation idempotent by recording or deriving a unique business request identifier.
5. Use an SQS FIFO queue with a meaningful MessageGroupId.

Correct Answer: C

 

Correct Answer

Answer C is correct because The retention period controls how long unconsumed messages remain available before SQS deletes them. This directly satisfies the decisive requirement in the scenario: message retention period.

Incorrect Answers

Answer A is incorrect because The visibility timeout keeps an in-flight message hidden from other consumers while it is being processed. It can be useful in other designs, but it does not satisfy the decisive requirement here: message retention period.

Answer B is incorrect because A DLQ captures messages that repeatedly fail so they do not circulate indefinitely in the source queue. It can be useful in other designs, but it does not satisfy the decisive requirement here: message retention period.

Answer D is incorrect because Idempotent processing prevents a redelivered message from causing a duplicate business side effect. It can be useful in other designs, but it does not satisfy the decisive requirement here: message retention period.

Answer E is incorrect because FIFO queues serialize messages within each message group while allowing other groups to progress. It can be useful in other designs, but it does not satisfy the decisive requirement here: message retention period.

 

Question 11

A checkout service can accept 20,000 orders in one minute, but a legacy fulfillment system can process only 500 per minute. Which pattern best protects the downstream system?

1. Use exponential backoff with jitter for retries.
2. Place an SQS queue between the producer and the downstream consumer.
3. Use backlog-aware scaling and throttle or shed producer load when backlog age exceeds the service objective.
4. Redrive repeatedly failing items to a DLQ.
5. Limit consumer concurrency to the safe downstream capacity.

Correct Answer: B

 

Correct Answer

Answer B is correct because A durable queue absorbs bursts and lets consumers drain work at a safe rate. This directly satisfies the decisive requirement in the scenario: buffer demand in a queue.

Incorrect Answers

Answer A is incorrect because Backoff with jitter spreads retries over time and reduces a synchronized retry storm against a recovering dependency. It can be useful in other designs, but it does not satisfy the decisive requirement here: buffer demand in a queue.

Answer C is incorrect because Backpressure requires the system to react when demand persistently exceeds safe processing capacity. It can be useful in other designs, but it does not satisfy the decisive requirement here: buffer demand in a queue.

Answer D is incorrect because A DLQ isolates poison messages from healthy work after a bounded number of receives. It can be useful in other designs, but it does not satisfy the decisive requirement here: buffer demand in a queue.

Answer E is incorrect because A concurrency cap prevents workers from exceeding a known dependency limit. It can be useful in other designs, but it does not satisfy the decisive requirement here: buffer demand in a queue.

 

Question 12

An SQS-triggered Lambda consumer can overwhelm a database when thousands of messages arrive together. Which control should be applied?

1. Place an SQS queue between the producer and the downstream consumer.
2. Redrive repeatedly failing items to a DLQ.
3. Use backlog-aware scaling and throttle or shed producer load when backlog age exceeds the service objective.
4. Limit consumer concurrency to the safe downstream capacity.
5. Use exponential backoff with jitter for retries.

Correct Answer: D

 

Correct Answer

Answer D is correct because A concurrency cap prevents workers from exceeding a known dependency limit. This directly satisfies the decisive requirement in the scenario: cap consumer concurrency.

Incorrect Answers

Answer A is incorrect because A durable queue absorbs bursts and lets consumers drain work at a safe rate. It can be useful in other designs, but it does not satisfy the decisive requirement here: cap consumer concurrency.

Answer B is incorrect because A DLQ isolates poison messages from healthy work after a bounded number of receives. It can be useful in other designs, but it does not satisfy the decisive requirement here: cap consumer concurrency.

Answer C is incorrect because Backpressure requires the system to react when demand persistently exceeds safe processing capacity. It can be useful in other designs, but it does not satisfy the decisive requirement here: cap consumer concurrency.

Answer E is incorrect because Backoff with jitter spreads retries over time and reduces a synchronized retry storm against a recovering dependency. It can be useful in other designs, but it does not satisfy the decisive requirement here: cap consumer concurrency.

 

Question 13

A single malformed order keeps failing and consuming worker time while good orders are waiting. Which mechanism should be used?

1. Redrive repeatedly failing items to a DLQ.
2. Place an SQS queue between the producer and the downstream consumer.
3. Limit consumer concurrency to the safe downstream capacity.
4. Use backlog-aware scaling and throttle or shed producer load when backlog age exceeds the service objective.
5. Use exponential backoff with jitter for retries.

Correct Answer: A

 

Correct Answer

Answer A is correct because A DLQ isolates poison messages from healthy work after a bounded number of receives. This directly satisfies the decisive requirement in the scenario: redrive failures to DLQ.

Incorrect Answers

Answer B is incorrect because A durable queue absorbs bursts and lets consumers drain work at a safe rate. It can be useful in other designs, but it does not satisfy the decisive requirement here: redrive failures to DLQ.

Answer C is incorrect because A concurrency cap prevents workers from exceeding a known dependency limit. It can be useful in other designs, but it does not satisfy the decisive requirement here: redrive failures to DLQ.

Answer D is incorrect because Backpressure requires the system to react when demand persistently exceeds safe processing capacity. It can be useful in other designs, but it does not satisfy the decisive requirement here: redrive failures to DLQ.

Answer E is incorrect because Backoff with jitter spreads retries over time and reduces a synchronized retry storm against a recovering dependency. It can be useful in other designs, but it does not satisfy the decisive requirement here: redrive failures to DLQ.

 

Question 14

A producer is creating work faster than consumers can safely process it for hours. Which operational response best represents backpressure?

1. Limit consumer concurrency to the safe downstream capacity.
2. Use exponential backoff with jitter for retries.
3. Use backlog-aware scaling and throttle or shed producer load when backlog age exceeds the service objective.
4. Place an SQS queue between the producer and the downstream consumer.
5. Redrive repeatedly failing items to a DLQ.

Correct Answer: C

 

Correct Answer

Answer C is correct because Backpressure requires the system to react when demand persistently exceeds safe processing capacity. This directly satisfies the decisive requirement in the scenario: slow intake or scale consumers based on backlog.

Incorrect Answers

Answer A is incorrect because A concurrency cap prevents workers from exceeding a known dependency limit. It can be useful in other designs, but it does not satisfy the decisive requirement here: slow intake or scale consumers based on backlog.

Answer B is incorrect because Backoff with jitter spreads retries over time and reduces a synchronized retry storm against a recovering dependency. It can be useful in other designs, but it does not satisfy the decisive requirement here: slow intake or scale consumers based on backlog.

Answer D is incorrect because A durable queue absorbs bursts and lets consumers drain work at a safe rate. It can be useful in other designs, but it does not satisfy the decisive requirement here: slow intake or scale consumers based on backlog.

Answer E is incorrect because A DLQ isolates poison messages from healthy work after a bounded number of receives. It can be useful in other designs, but it does not satisfy the decisive requirement here: slow intake or scale consumers based on backlog.

 

Question 15

A dependency outage causes thousands of workers to retry immediately and makes recovery harder. Which retry design should replace the current behavior?

1. Use backlog-aware scaling and throttle or shed producer load when backlog age exceeds the service objective.
2. Limit consumer concurrency to the safe downstream capacity.
3. Place an SQS queue between the producer and the downstream consumer.
4. Redrive repeatedly failing items to a DLQ.
5. Use exponential backoff with jitter for retries.

Correct Answer: E

 

Correct Answer

Answer E is correct because Backoff with jitter spreads retries over time and reduces a synchronized retry storm against a recovering dependency. This directly satisfies the decisive requirement in the scenario: exponential backoff with jitter.

Incorrect Answers

Answer A is incorrect because Backpressure requires the system to react when demand persistently exceeds safe processing capacity. It can be useful in other designs, but it does not satisfy the decisive requirement here: exponential backoff with jitter.

Answer B is incorrect because A concurrency cap prevents workers from exceeding a known dependency limit. It can be useful in other designs, but it does not satisfy the decisive requirement here: exponential backoff with jitter.

Answer C is incorrect because A durable queue absorbs bursts and lets consumers drain work at a safe rate. It can be useful in other designs, but it does not satisfy the decisive requirement here: exponential backoff with jitter.

Answer D is incorrect because A DLQ isolates poison messages from healthy work after a bounded number of receives. It can be useful in other designs, but it does not satisfy the decisive requirement here: exponential backoff with jitter.

 

Question 16

An order must reserve inventory, then authorize payment, then create shipment only after both earlier steps succeed. Which Step Functions pattern best represents the dependency?

1. Define explicit compensating steps for previously completed side effects.
2. Use a Step Functions Parallel state.
3. Configure Retry and Catch behavior on the failing task state.
4. Chain states in a Step Functions workflow so each dependent step starts after the prior step succeeds.
5. Use a Step Functions Wait state.

Correct Answer: D

 

Correct Answer

Answer D is correct because Sequential states model ordered dependencies and pass state between steps. This directly satisfies the decisive requirement in the scenario: sequential states.

Incorrect Answers

Answer A is incorrect because Compensation reverses or neutralizes completed distributed actions when a later business step fails. It can be useful in other designs, but it does not satisfy the decisive requirement here: sequential states.

Answer B is incorrect because A Parallel state runs independent branches concurrently and can join them before continuing. It can be useful in other designs, but it does not satisfy the decisive requirement here: sequential states.

Answer C is incorrect because Task-level retry and catch settings define the retry boundary and failure path for that integration. It can be useful in other designs, but it does not satisfy the decisive requirement here: sequential states.

Answer E is incorrect because A Wait state pauses workflow progression for a time or timestamp without consuming a continuously running worker. It can be useful in other designs, but it does not satisfy the decisive requirement here: sequential states.

 

Question 17

A media workflow must generate a thumbnail, transcript, and metadata in parallel before publishing the asset. Which workflow construct best fits?

1. Use a Step Functions Wait state.
2. Define explicit compensating steps for previously completed side effects.
3. Configure Retry and Catch behavior on the failing task state.
4. Use a Step Functions Parallel state.
5. Chain states in a Step Functions workflow so each dependent step starts after the prior step succeeds.

Correct Answer: D

 

Correct Answer

Answer D is correct because A Parallel state runs independent branches concurrently and can join them before continuing. This directly satisfies the decisive requirement in the scenario: Parallel state.

Incorrect Answers

Answer A is incorrect because A Wait state pauses workflow progression for a time or timestamp without consuming a continuously running worker. It can be useful in other designs, but it does not satisfy the decisive requirement here: Parallel state.

Answer B is incorrect because Compensation reverses or neutralizes completed distributed actions when a later business step fails. It can be useful in other designs, but it does not satisfy the decisive requirement here: Parallel state.

Answer C is incorrect because Task-level retry and catch settings define the retry boundary and failure path for that integration. It can be useful in other designs, but it does not satisfy the decisive requirement here: Parallel state.

Answer E is incorrect because Sequential states model ordered dependencies and pass state between steps. It can be useful in other designs, but it does not satisfy the decisive requirement here: Parallel state.

 

Question 18

A fraud review must pause for up to 24 hours before checking an external decision. Which workflow capability should model the delay without busy-waiting compute?

1. Define explicit compensating steps for previously completed side effects.
2. Use a Step Functions Parallel state.
3. Use a Step Functions Wait state.
4. Chain states in a Step Functions workflow so each dependent step starts after the prior step succeeds.
5. Configure Retry and Catch behavior on the failing task state.

Correct Answer: C

 

Correct Answer

Answer C is correct because A Wait state pauses workflow progression for a time or timestamp without consuming a continuously running worker. This directly satisfies the decisive requirement in the scenario: Wait state.

Incorrect Answers

Answer A is incorrect because Compensation reverses or neutralizes completed distributed actions when a later business step fails. It can be useful in other designs, but it does not satisfy the decisive requirement here: Wait state.

Answer B is incorrect because A Parallel state runs independent branches concurrently and can join them before continuing. It can be useful in other designs, but it does not satisfy the decisive requirement here: Wait state.

Answer D is incorrect because Sequential states model ordered dependencies and pass state between steps. It can be useful in other designs, but it does not satisfy the decisive requirement here: Wait state.

Answer E is incorrect because Task-level retry and catch settings define the retry boundary and failure path for that integration. It can be useful in other designs, but it does not satisfy the decisive requirement here: Wait state.

 

Question 19

A travel booking workflow reserves a hotel and flight. If car rental fails, the first two reservations must be canceled. Which design best handles the business rollback?

1. Use a Step Functions Parallel state.
2. Use a Step Functions Wait state.
3. Define explicit compensating steps for previously completed side effects.
4. Chain states in a Step Functions workflow so each dependent step starts after the prior step succeeds.
5. Configure Retry and Catch behavior on the failing task state.

Correct Answer: C

 

Correct Answer

Answer C is correct because Compensation reverses or neutralizes completed distributed actions when a later business step fails. This directly satisfies the decisive requirement in the scenario: compensating actions.

Incorrect Answers

Answer A is incorrect because A Parallel state runs independent branches concurrently and can join them before continuing. It can be useful in other designs, but it does not satisfy the decisive requirement here: compensating actions.

Answer B is incorrect because A Wait state pauses workflow progression for a time or timestamp without consuming a continuously running worker. It can be useful in other designs, but it does not satisfy the decisive requirement here: compensating actions.

Answer D is incorrect because Sequential states model ordered dependencies and pass state between steps. It can be useful in other designs, but it does not satisfy the decisive requirement here: compensating actions.

Answer E is incorrect because Task-level retry and catch settings define the retry boundary and failure path for that integration. It can be useful in other designs, but it does not satisfy the decisive requirement here: compensating actions.

 

Question 20

One API call in a state machine is temporarily unavailable. Which configuration should control retries only around that task?

1. Define explicit compensating steps for previously completed side effects.
2. Use a Step Functions Parallel state.
3. Chain states in a Step Functions workflow so each dependent step starts after the prior step succeeds.
4. Configure Retry and Catch behavior on the failing task state.
5. Use a Step Functions Wait state.

Correct Answer: D

 

Correct Answer

Answer D is correct because Task-level retry and catch settings define the retry boundary and failure path for that integration. This directly satisfies the decisive requirement in the scenario: task-level Retry/Catch policy.

Incorrect Answers

Answer A is incorrect because Compensation reverses or neutralizes completed distributed actions when a later business step fails. It can be useful in other designs, but it does not satisfy the decisive requirement here: task-level Retry/Catch policy.

Answer B is incorrect because A Parallel state runs independent branches concurrently and can join them before continuing. It can be useful in other designs, but it does not satisfy the decisive requirement here: task-level Retry/Catch policy.

Answer C is incorrect because Sequential states model ordered dependencies and pass state between steps. It can be useful in other designs, but it does not satisfy the decisive requirement here: task-level Retry/Catch policy.

Answer E is incorrect because A Wait state pauses workflow progression for a time or timestamp without consuming a continuously running worker. It can be useful in other designs, but it does not satisfy the decisive requirement here: task-level Retry/Catch policy.

 

Question 21

An order event must be processed independently by shipping, analytics, and fraud systems, and one slow consumer must not block the others. Which design is best?

1. Use EventBridge archive and replay for events that need controlled reprocessing.
2. Give each consumer its own durable queue rather than sharing one queue.
3. Use EventBridge rules that match only the event patterns needed by each target.
4. Publish to SNS and subscribe a separate SQS queue for each durable consumer.
5. Use an SQS queue dedicated to the consumer that needs durable backlog.

Correct Answer: D

 

Correct Answer

Answer D is correct because SNS can fan out one publication while each SQS subscription maintains independent durable backlog and retry behavior. This directly satisfies the decisive requirement in the scenario: fan out to separate queues.

Incorrect Answers

Answer A is incorrect because An EventBridge archive can retain matching events and later replay them to an event bus. It can be useful in other designs, but it does not satisfy the decisive requirement here: fan out to separate queues.

Answer B is incorrect because Separate queues isolate backlog, visibility timeout, redrive, and failure handling per consumer. It can be useful in other designs, but it does not satisfy the decisive requirement here: fan out to separate queues.

Answer C is incorrect because EventBridge rules route events by content so consumers need not receive unrelated event types. It can be useful in other designs, but it does not satisfy the decisive requirement here: fan out to separate queues.

Answer E is incorrect because A queue persists work for a temporarily unavailable consumer until it can resume processing. It can be useful in other designs, but it does not satisfy the decisive requirement here: fan out to separate queues.

 

Question 22

A central event bus receives many event types, but a payroll consumer must receive only employee-compensation events. Which capability should be used?

1. Use EventBridge rules that match only the event patterns needed by each target.
2. Publish to SNS and subscribe a separate SQS queue for each durable consumer.
3. Give each consumer its own durable queue rather than sharing one queue.
4. Use EventBridge archive and replay for events that need controlled reprocessing.
5. Use an SQS queue dedicated to the consumer that needs durable backlog.

Correct Answer: A

 

Correct Answer

Answer A is correct because EventBridge rules route events by content so consumers need not receive unrelated event types. This directly satisfies the decisive requirement in the scenario: rule-based event filtering.

Incorrect Answers

Answer B is incorrect because SNS can fan out one publication while each SQS subscription maintains independent durable backlog and retry behavior. It can be useful in other designs, but it does not satisfy the decisive requirement here: rule-based event filtering.

Answer C is incorrect because Separate queues isolate backlog, visibility timeout, redrive, and failure handling per consumer. It can be useful in other designs, but it does not satisfy the decisive requirement here: rule-based event filtering.

Answer D is incorrect because An EventBridge archive can retain matching events and later replay them to an event bus. It can be useful in other designs, but it does not satisfy the decisive requirement here: rule-based event filtering.

Answer E is incorrect because A queue persists work for a temporarily unavailable consumer until it can resume processing. It can be useful in other designs, but it does not satisfy the decisive requirement here: rule-based event filtering.

 

Question 23

Developers need to replay selected historical events after deploying a corrected consumer. Which capability best supports this?

1. Use EventBridge archive and replay for events that need controlled reprocessing.
2. Publish to SNS and subscribe a separate SQS queue for each durable consumer.
3. Use an SQS queue dedicated to the consumer that needs durable backlog.
4. Use EventBridge rules that match only the event patterns needed by each target.
5. Give each consumer its own durable queue rather than sharing one queue.

Correct Answer: A

 

Correct Answer

Answer A is correct because An EventBridge archive can retain matching events and later replay them to an event bus. This directly satisfies the decisive requirement in the scenario: event archive and replay.

Incorrect Answers

Answer B is incorrect because SNS can fan out one publication while each SQS subscription maintains independent durable backlog and retry behavior. It can be useful in other designs, but it does not satisfy the decisive requirement here: event archive and replay.

Answer C is incorrect because A queue persists work for a temporarily unavailable consumer until it can resume processing. It can be useful in other designs, but it does not satisfy the decisive requirement here: event archive and replay.

Answer D is incorrect because EventBridge rules route events by content so consumers need not receive unrelated event types. It can be useful in other designs, but it does not satisfy the decisive requirement here: event archive and replay.

Answer E is incorrect because Separate queues isolate backlog, visibility timeout, redrive, and failure handling per consumer. It can be useful in other designs, but it does not satisfy the decisive requirement here: event archive and replay.

 

Question 24

A consumer may be offline for several hours but must process every event after it returns. Which component should retain that consumer’s backlog?

1. Use EventBridge rules that match only the event patterns needed by each target.
2. Publish to SNS and subscribe a separate SQS queue for each durable consumer.
3. Use an SQS queue dedicated to the consumer that needs durable backlog.
4. Give each consumer its own durable queue rather than sharing one queue.
5. Use EventBridge archive and replay for events that need controlled reprocessing.

Correct Answer: C

 

Correct Answer

Answer C is correct because A queue persists work for a temporarily unavailable consumer until it can resume processing. This directly satisfies the decisive requirement in the scenario: dedicated durable queue.

Incorrect Answers

Answer A is incorrect because EventBridge rules route events by content so consumers need not receive unrelated event types. It can be useful in other designs, but it does not satisfy the decisive requirement here: dedicated durable queue.

Answer B is incorrect because SNS can fan out one publication while each SQS subscription maintains independent durable backlog and retry behavior. It can be useful in other designs, but it does not satisfy the decisive requirement here: dedicated durable queue.

Answer D is incorrect because Separate queues isolate backlog, visibility timeout, redrive, and failure handling per consumer. It can be useful in other designs, but it does not satisfy the decisive requirement here: dedicated durable queue.

Answer E is incorrect because An EventBridge archive can retain matching events and later replay them to an event bus. It can be useful in other designs, but it does not satisfy the decisive requirement here: dedicated durable queue.

 

Question 25

Five consumers subscribe to the same event, and failures in one consumer must not cause redelivery to the other four. Which design best isolates failure handling?

1. Publish to SNS and subscribe a separate SQS queue for each durable consumer.
2. Use an SQS queue dedicated to the consumer that needs durable backlog.
3. Give each consumer its own durable queue rather than sharing one queue.
4. Use EventBridge rules that match only the event patterns needed by each target.
5. Use EventBridge archive and replay for events that need controlled reprocessing.

Correct Answer: C

 

Correct Answer

Answer C is correct because Separate queues isolate backlog, visibility timeout, redrive, and failure handling per consumer. This directly satisfies the decisive requirement in the scenario: one queue per consumer.

Incorrect Answers

Answer A is incorrect because SNS can fan out one publication while each SQS subscription maintains independent durable backlog and retry behavior. It can be useful in other designs, but it does not satisfy the decisive requirement here: one queue per consumer.

Answer B is incorrect because A queue persists work for a temporarily unavailable consumer until it can resume processing. It can be useful in other designs, but it does not satisfy the decisive requirement here: one queue per consumer.

Answer D is incorrect because EventBridge rules route events by content so consumers need not receive unrelated event types. It can be useful in other designs, but it does not satisfy the decisive requirement here: one queue per consumer.

Answer E is incorrect because An EventBridge archive can retain matching events and later replay them to an event bus. It can be useful in other designs, but it does not satisfy the decisive requirement here: one queue per consumer.

 

Question 26

A client submits a report job that takes 20 minutes. The API must return quickly instead of holding the HTTP connection. Which interface pattern should be used?

1. Persist job state by job ID in a durable store and expose a status endpoint.
2. Configure API Gateway throttling and quotas to bound accepted request rates.
3. Publish a completion event that a callback or notification component delivers to the client.
4. Use a Step Functions Standard workflow for the long-running orchestration.
5. Return an accepted response with a job ID and enqueue the work for asynchronous processing.

Correct Answer: E

 

Correct Answer

Answer E is correct because An asynchronous submission pattern decouples request acknowledgement from completion of long-running work. This directly satisfies the decisive requirement in the scenario: return job identifier and queue work.

Incorrect Answers

Answer A is incorrect because A status record lets clients retrieve progress and terminal results without holding the original request open. It can be useful in other designs, but it does not satisfy the decisive requirement here: return job identifier and queue work.

Answer B is incorrect because API throttling limits inbound demand before it can overwhelm the downstream service. It can be useful in other designs, but it does not satisfy the decisive requirement here: return job identifier and queue work.

Answer C is incorrect because A completion event lets the system notify an interested client after asynchronous work reaches a terminal state. It can be useful in other designs, but it does not satisfy the decisive requirement here: return job identifier and queue work.

Answer D is incorrect because Standard workflows persist state and coordinate long-running, retriable business processes. It can be useful in other designs, but it does not satisfy the decisive requirement here: return job identifier and queue work.

 

Question 27

After submitting an asynchronous export, clients need to query progress by job ID. Which design best supports result lookup?

1. Use a Step Functions Standard workflow for the long-running orchestration.
2. Configure API Gateway throttling and quotas to bound accepted request rates.
3. Persist job state by job ID in a durable store and expose a status endpoint.
4. Publish a completion event that a callback or notification component delivers to the client.
5. Return an accepted response with a job ID and enqueue the work for asynchronous processing.

Correct Answer: C

 

Correct Answer

Answer C is correct because A status record lets clients retrieve progress and terminal results without holding the original request open. This directly satisfies the decisive requirement in the scenario: persist status keyed by job ID.

Incorrect Answers

Answer A is incorrect because Standard workflows persist state and coordinate long-running, retriable business processes. It can be useful in other designs, but it does not satisfy the decisive requirement here: persist status keyed by job ID.

Answer B is incorrect because API throttling limits inbound demand before it can overwhelm the downstream service. It can be useful in other designs, but it does not satisfy the decisive requirement here: persist status keyed by job ID.

Answer D is incorrect because A completion event lets the system notify an interested client after asynchronous work reaches a terminal state. It can be useful in other designs, but it does not satisfy the decisive requirement here: persist status keyed by job ID.

Answer E is incorrect because An asynchronous submission pattern decouples request acknowledgement from completion of long-running work. It can be useful in other designs, but it does not satisfy the decisive requirement here: persist status keyed by job ID.

 

Question 28

A business process can run for hours and spans several retriable service calls. Which managed service should coordinate the long-running state?

1. Persist job state by job ID in a durable store and expose a status endpoint.
2. Configure API Gateway throttling and quotas to bound accepted request rates.
3. Publish a completion event that a callback or notification component delivers to the client.
4. Use a Step Functions Standard workflow for the long-running orchestration.
5. Return an accepted response with a job ID and enqueue the work for asynchronous processing.

Correct Answer: D

 

Correct Answer

Answer D is correct because Standard workflows persist state and coordinate long-running, retriable business processes. This directly satisfies the decisive requirement in the scenario: Step Functions Standard workflow.

Incorrect Answers

Answer A is incorrect because A status record lets clients retrieve progress and terminal results without holding the original request open. It can be useful in other designs, but it does not satisfy the decisive requirement here: Step Functions Standard workflow.

Answer B is incorrect because API throttling limits inbound demand before it can overwhelm the downstream service. It can be useful in other designs, but it does not satisfy the decisive requirement here: Step Functions Standard workflow.

Answer C is incorrect because A completion event lets the system notify an interested client after asynchronous work reaches a terminal state. It can be useful in other designs, but it does not satisfy the decisive requirement here: Step Functions Standard workflow.

Answer E is incorrect because An asynchronous submission pattern decouples request acknowledgement from completion of long-running work. It can be useful in other designs, but it does not satisfy the decisive requirement here: Step Functions Standard workflow.

 

Question 29

A customer wants a notification when an asynchronous job finishes instead of polling continuously. Which pattern is best?

1. Use a Step Functions Standard workflow for the long-running orchestration.
2. Publish a completion event that a callback or notification component delivers to the client.
3. Configure API Gateway throttling and quotas to bound accepted request rates.
4. Persist job state by job ID in a durable store and expose a status endpoint.
5. Return an accepted response with a job ID and enqueue the work for asynchronous processing.

Correct Answer: B

 

Correct Answer

Answer B is correct because A completion event lets the system notify an interested client after asynchronous work reaches a terminal state. This directly satisfies the decisive requirement in the scenario: publish completion event to callback worker.

Incorrect Answers

Answer A is incorrect because Standard workflows persist state and coordinate long-running, retriable business processes. It can be useful in other designs, but it does not satisfy the decisive requirement here: publish completion event to callback worker.

Answer C is incorrect because API throttling limits inbound demand before it can overwhelm the downstream service. It can be useful in other designs, but it does not satisfy the decisive requirement here: publish completion event to callback worker.

Answer D is incorrect because A status record lets clients retrieve progress and terminal results without holding the original request open. It can be useful in other designs, but it does not satisfy the decisive requirement here: publish completion event to callback worker.

Answer E is incorrect because An asynchronous submission pattern decouples request acknowledgement from completion of long-running work. It can be useful in other designs, but it does not satisfy the decisive requirement here: publish completion event to callback worker.

 

Question 30

A public API is receiving more requests than the downstream service can safely accept. Which API control should protect the backend?

1. Use a Step Functions Standard workflow for the long-running orchestration.
2. Configure API Gateway throttling and quotas to bound accepted request rates.
3. Return an accepted response with a job ID and enqueue the work for asynchronous processing.
4. Publish a completion event that a callback or notification component delivers to the client.
5. Persist job state by job ID in a durable store and expose a status endpoint.

Correct Answer: B

 

Correct Answer

Answer B is correct because API throttling limits inbound demand before it can overwhelm the downstream service. This directly satisfies the decisive requirement in the scenario: API Gateway throttling.

Incorrect Answers

Answer A is incorrect because Standard workflows persist state and coordinate long-running, retriable business processes. It can be useful in other designs, but it does not satisfy the decisive requirement here: API Gateway throttling.

Answer C is incorrect because An asynchronous submission pattern decouples request acknowledgement from completion of long-running work. It can be useful in other designs, but it does not satisfy the decisive requirement here: API Gateway throttling.

Answer D is incorrect because A completion event lets the system notify an interested client after asynchronous work reaches a terminal state. It can be useful in other designs, but it does not satisfy the decisive requirement here: API Gateway throttling.

Answer E is incorrect because A status record lets clients retrieve progress and terminal results without holding the original request open. It can be useful in other designs, but it does not satisfy the decisive requirement here: API Gateway throttling.

 

Question 31

A monolith has a slow invoice-generation function that can be extracted without changing the order transaction. Which seam gives the new worker an independent failure boundary?

1. Use Amazon MQ for compatible broker protocols during the migration.
2. Publish a stable domain event through EventBridge.
3. Implement idempotency around the business side effect using a stable request or transaction key.
4. Insert an SQS queue at the asynchronous work boundary.
5. Replace direct shared-table updates with an owned service API or event contract.

Correct Answer: D

 

Correct Answer

Answer D is correct because A queue provides a durable seam that separates the producer transaction from independently scaled workers. This directly satisfies the decisive requirement in the scenario: queue invoice work.

Incorrect Answers

Answer A is incorrect because Amazon MQ can preserve established messaging protocols while applications are migrated incrementally. It can be useful in other designs, but it does not satisfy the decisive requirement here: queue invoice work.

Answer B is incorrect because An event contract lets additional consumers subscribe without adding point-to-point producer dependencies. It can be useful in other designs, but it does not satisfy the decisive requirement here: queue invoice work.

Answer C is incorrect because Idempotency allows a retried request to return the existing result instead of repeating a destructive side effect. It can be useful in other designs, but it does not satisfy the decisive requirement here: queue invoice work.

Answer E is incorrect because Independent data ownership removes the shared database as a deployment and failure coupling point. It can be useful in other designs, but it does not satisfy the decisive requirement here: queue invoice work.

 

Question 32

Several new services need to react to business events from a legacy application without direct point-to-point integrations. Which seam should be introduced?

1. Insert an SQS queue at the asynchronous work boundary.
2. Use Amazon MQ for compatible broker protocols during the migration.
3. Replace direct shared-table updates with an owned service API or event contract.
4. Implement idempotency around the business side effect using a stable request or transaction key.
5. Publish a stable domain event through EventBridge.

Correct Answer: E

 

Correct Answer

Answer E is correct because An event contract lets additional consumers subscribe without adding point-to-point producer dependencies. This directly satisfies the decisive requirement in the scenario: publish domain events.

Incorrect Answers

Answer A is incorrect because A queue provides a durable seam that separates the producer transaction from independently scaled workers. It can be useful in other designs, but it does not satisfy the decisive requirement here: publish domain events.

Answer B is incorrect because Amazon MQ can preserve established messaging protocols while applications are migrated incrementally. It can be useful in other designs, but it does not satisfy the decisive requirement here: publish domain events.

Answer C is incorrect because Independent data ownership removes the shared database as a deployment and failure coupling point. It can be useful in other designs, but it does not satisfy the decisive requirement here: publish domain events.

Answer D is incorrect because Idempotency allows a retried request to return the existing result instead of repeating a destructive side effect. It can be useful in other designs, but it does not satisfy the decisive requirement here: publish domain events.

 

Question 33

Two services currently update the same database tables and deployments frequently break each other. Which migration goal is most important?

1. Publish a stable domain event through EventBridge.
2. Use Amazon MQ for compatible broker protocols during the migration.
3. Implement idempotency around the business side effect using a stable request or transaction key.
4. Replace direct shared-table updates with an owned service API or event contract.
5. Insert an SQS queue at the asynchronous work boundary.

Correct Answer: D

 

Correct Answer

Answer D is correct because Independent data ownership removes the shared database as a deployment and failure coupling point. This directly satisfies the decisive requirement in the scenario: replace table sharing with service contract.

Incorrect Answers

Answer A is incorrect because An event contract lets additional consumers subscribe without adding point-to-point producer dependencies. It can be useful in other designs, but it does not satisfy the decisive requirement here: replace table sharing with service contract.

Answer B is incorrect because Amazon MQ can preserve established messaging protocols while applications are migrated incrementally. It can be useful in other designs, but it does not satisfy the decisive requirement here: replace table sharing with service contract.

Answer C is incorrect because Idempotency allows a retried request to return the existing result instead of repeating a destructive side effect. It can be useful in other designs, but it does not satisfy the decisive requirement here: replace table sharing with service contract.

Answer E is incorrect because A queue provides a durable seam that separates the producer transaction from independently scaled workers. It can be useful in other designs, but it does not satisfy the decisive requirement here: replace table sharing with service contract.

 

Question 34

A legacy system must move to AWS but uses AMQP/JMS interfaces that cannot yet be rewritten. Which managed service provides the least disruptive messaging seam?

1. Publish a stable domain event through EventBridge.
2. Insert an SQS queue at the asynchronous work boundary.
3. Replace direct shared-table updates with an owned service API or event contract.
4. Use Amazon MQ for compatible broker protocols during the migration.
5. Implement idempotency around the business side effect using a stable request or transaction key.

Correct Answer: D

 

Correct Answer

Answer D is correct because Amazon MQ can preserve established messaging protocols while applications are migrated incrementally. This directly satisfies the decisive requirement in the scenario: managed compatible broker.

Incorrect Answers

Answer A is incorrect because An event contract lets additional consumers subscribe without adding point-to-point producer dependencies. It can be useful in other designs, but it does not satisfy the decisive requirement here: managed compatible broker.

Answer B is incorrect because A queue provides a durable seam that separates the producer transaction from independently scaled workers. It can be useful in other designs, but it does not satisfy the decisive requirement here: managed compatible broker.

Answer C is incorrect because Independent data ownership removes the shared database as a deployment and failure coupling point. It can be useful in other designs, but it does not satisfy the decisive requirement here: managed compatible broker.

Answer E is incorrect because Idempotency allows a retried request to return the existing result instead of repeating a destructive side effect. It can be useful in other designs, but it does not satisfy the decisive requirement here: managed compatible broker.

 

Question 35

A newly extracted payment side effect may be retried after timeouts. Which property must the new service implement before retries are safe?

1. Publish a stable domain event through EventBridge.
2. Use Amazon MQ for compatible broker protocols during the migration.
3. Insert an SQS queue at the asynchronous work boundary.
4. Replace direct shared-table updates with an owned service API or event contract.
5. Implement idempotency around the business side effect using a stable request or transaction key.

Correct Answer: E

 

Correct Answer

Answer E is correct because Idempotency allows a retried request to return the existing result instead of repeating a destructive side effect. This directly satisfies the decisive requirement in the scenario: idempotent business operation.

Incorrect Answers

Answer A is incorrect because An event contract lets additional consumers subscribe without adding point-to-point producer dependencies. It can be useful in other designs, but it does not satisfy the decisive requirement here: idempotent business operation.

Answer B is incorrect because Amazon MQ can preserve established messaging protocols while applications are migrated incrementally. It can be useful in other designs, but it does not satisfy the decisive requirement here: idempotent business operation.

Answer C is incorrect because A queue provides a durable seam that separates the producer transaction from independently scaled workers. It can be useful in other designs, but it does not satisfy the decisive requirement here: idempotent business operation.

Answer D is incorrect because Independent data ownership removes the shared database as a deployment and failure coupling point. It can be useful in other designs, but it does not satisfy the decisive requirement here: idempotent business operation.