Amazon AWS Certified Solutions Architect - Associate SAA-C03 Practice Test 34
Question 330
One tenant's hot partition key dominates a Kinesis stream shard and throttles writes. What should be revised?
- Distribute partition keys more evenly across available shards while preserving required ordering groups.
- Persist records in a retained Kinesis Data Stream and process them when consumers return.
- Choose the customer ID as an appropriate Kinesis partition key with sufficient shard distribution.
- Increase Kinesis stream retention to cover the required replay window.
Explanation
Correct Answer
A:Partition-key hashing determines shard load; a hot key cannot be fixed by adding consumers alone. Measure per-shard incoming records and compare after repartitioning.
Incorrect Answers
B:A stream buffers records independently of consumer availability within retention limits. Distribute partition keys more evenly across available shards while preserving required ordering groups. Measure per-shard incoming records and compare after repartitioning.
C:Records sharing a partition key reach the same shard and preserve per-shard order; overall global order is not guaranteed. Distribute partition keys more evenly across available shards while preserving required ordering groups. Measure per-shard incoming records and compare after repartitioning.
D:Kinesis retains records for a configured interval so recovery depends on that window. Distribute partition keys more evenly across available shards while preserving required ordering groups. Measure per-shard incoming records and compare after repartitioning.
Question 331
A stream consumer lags beyond the default retention time during an outage. What must be adjusted?
- Persist records in a retained Kinesis Data Stream and process them when consumers return.
- Choose the customer ID as an appropriate Kinesis partition key with sufficient shard distribution.
- Increase Kinesis stream retention to cover the required replay window.
- Use Kinesis Data Streams with separate consumer applications.
Explanation
Correct Answer
C:Kinesis retains records for a configured interval so recovery depends on that window. Pause a consumer for a bounded period and verify replay from its checkpoint.
Incorrect Answers
A:A stream buffers records independently of consumer availability within retention limits. Increase Kinesis stream retention to cover the required replay window. Pause a consumer for a bounded period and verify replay from its checkpoint.
B:Records sharing a partition key reach the same shard and preserve per-shard order; overall global order is not guaranteed. Increase Kinesis stream retention to cover the required replay window. Pause a consumer for a bounded period and verify replay from its checkpoint.
D:A retained append-only stream lets multiple consumers process the event sequence at independent rates. Increase Kinesis stream retention to cover the required replay window. Pause a consumer for a bounded period and verify replay from its checkpoint.
Question 332
An analytics application needs dedicated stream read throughput not shared with another consumer. Which mode should be considered?
- Choose the customer ID as an appropriate Kinesis partition key with sufficient shard distribution.
- Use Kinesis enhanced fan-out for the dedicated registered consumer.
- Increase Kinesis stream retention to cover the required replay window.
- Use Kinesis Data Streams with separate consumer applications.
Explanation
Correct Answer
B:Enhanced fan-out provides consumer-specific throughput paths rather than sharing ordinary shard read capacity. Measure read latency under two independent consumers before and after enabling it.
Incorrect Answers
A:Records sharing a partition key reach the same shard and preserve per-shard order; overall global order is not guaranteed. Use Kinesis enhanced fan-out for the dedicated registered consumer. Measure read latency under two independent consumers before and after enabling it.
C:Kinesis retains records for a configured interval so recovery depends on that window. Use Kinesis enhanced fan-out for the dedicated registered consumer. Measure read latency under two independent consumers before and after enabling it.
D:A retained append-only stream lets multiple consumers process the event sequence at independent rates. Use Kinesis enhanced fan-out for the dedicated registered consumer. Measure read latency under two independent consumers before and after enabling it.
Question 333
A streaming application needs ordering for events of one customer but independent processing across customers. Which partitioning strategy fits?
- Use Kinesis enhanced fan-out for the dedicated registered consumer.
- Choose the customer ID as an appropriate Kinesis partition key with sufficient shard distribution.
- Distribute partition keys more evenly across available shards while preserving required ordering groups.
- Increase Kinesis stream retention to cover the required replay window.
Explanation
Correct Answer
B:Records sharing a partition key reach the same shard and preserve per-shard order; overall global order is not guaranteed. Send interleaved customer events and inspect per-customer ordering.
Incorrect Answers
A:Enhanced fan-out provides consumer-specific throughput paths rather than sharing ordinary shard read capacity. Choose the customer ID as an appropriate Kinesis partition key with sufficient shard distribution. Send interleaved customer events and inspect per-customer ordering.
C:Partition-key hashing determines shard load; a hot key cannot be fixed by adding consumers alone. Choose the customer ID as an appropriate Kinesis partition key with sufficient shard distribution. Send interleaved customer events and inspect per-customer ordering.
D:Kinesis retains records for a configured interval so recovery depends on that window. Choose the customer ID as an appropriate Kinesis partition key with sufficient shard distribution. Send interleaved customer events and inspect per-customer ordering.
Question 334
A producer must avoid data loss when no consumers are running overnight. What service behavior is needed?
- Shorten the stream retention period so records expire faster while consumers are offline.
- Use Amazon Data Firehose to deliver supported streaming source records to S3.
- Use Kinesis Data Streams with separate consumer applications.
- Persist records in a retained Kinesis Data Stream and process them when consumers return.
Explanation
Correct Answer
D:A stream buffers records independently of consumer availability within retention limits. Stop consumers and verify incoming records remain readable later.
Incorrect Answers
A:Reducing retention makes unread records expire sooner and increases the risk of data loss during overnight consumer downtime. Keep the Kinesis stream retention longer than the expected outage and resume processing from retained records.
B:Data Firehose buffers and delivers records to destinations such as S3 with managed scaling. Persist records in a retained Kinesis Data Stream and process them when consumers return. Stop consumers and verify incoming records remain readable later.
C:A retained append-only stream lets multiple consumers process the event sequence at independent rates. Persist records in a retained Kinesis Data Stream and process them when consumers return. Stop consumers and verify incoming records remain readable later.
Question 335
A Lambda consumer repeatedly reprocesses stream records after a failed batch. What application behavior must be safe?
- Implement idempotent downstream writes and checkpoints compatible with stream retries.
- Persist records in a retained Kinesis Data Stream and process them when consumers return.
- Increase Kinesis stream retention to cover the required replay window.
- Use Kinesis Data Streams with separate consumer applications.
Explanation
Correct Answer
A:A failed batch can be retried, so consumers must tolerate duplicate deliveries rather than rely on a strict once-only side effect. Force one downstream failure then retry and inspect effective writes.
Incorrect Answers
B:A stream buffers records independently of consumer availability within retention limits. Implement idempotent downstream writes and checkpoints compatible with stream retries. Force one downstream failure then retry and inspect effective writes.
C:Kinesis retains records for a configured interval so recovery depends on that window. Implement idempotent downstream writes and checkpoints compatible with stream retries. Force one downstream failure then retry and inspect effective writes.
D:A retained append-only stream lets multiple consumers process the event sequence at independent rates. Implement idempotent downstream writes and checkpoints compatible with stream retries. Force one downstream failure then retry and inspect effective writes.
Question 336
The business needs managed delivery of streaming records to S3 without writing a bespoke ingestion consumer. Which service is most suitable?
- Persist records in a retained Kinesis Data Stream and process them when consumers return.
- Use Kinesis Data Streams with separate consumer applications.
- Distribute partition keys more evenly across available shards while preserving required ordering groups.
- Use Amazon Data Firehose to deliver supported streaming source records to S3.
Explanation
Correct Answer
D:Data Firehose buffers and delivers records to destinations such as S3 with managed scaling. Send test stream records and inspect S3 objects after the configured buffering period.
Incorrect Answers
A:A stream buffers records independently of consumer availability within retention limits. Use Amazon Data Firehose to deliver supported streaming source records to S3. Send test stream records and inspect S3 objects after the configured buffering period.
B:A retained append-only stream lets multiple consumers process the event sequence at independent rates. Use Amazon Data Firehose to deliver supported streaming source records to S3. Send test stream records and inspect S3 objects after the configured buffering period.
C:Partition-key hashing determines shard load; a hot key cannot be fixed by adding consumers alone. Data Firehose buffers and delivers records to destinations such as S3 with managed scaling.
Question 337
A central audit account must collect supported operational events from member accounts and route a subset to an analysis Lambda. Events are being rejected at the central bus, and the receiving team wants recoverable processing failures. Which THREE controls establish the complete path? (Select THREE.)
- Create member-account EventBridge rules that forward the selected events to the central event-bus ARN.
- Configure only a CloudWatch dashboard with no rules or bus permissions.
- Use an S3 lifecycle transition rule as the sole mechanism to invoke EventBridge targets.
- Disable the central bus resource policy entirely and assume source rules can always publish.
- Grant the intended member principals or organizations permission in the central EventBridge event-bus resource policy.
- Set a valid rule or target retry and dead-letter policy for the central analysis Lambda integration.
Explanation
Correct Answers
A:A bus policy permits receipt but does not itself send source events; member rules must forward the selected event patterns.
E:Cross-account PutEvents requires the receiving bus to authorize the appropriate sending source.
F:Target retry and DLQ capture eligible failures so events are not silently discarded after retry exhaustion.
Incorrect Answers
B:Monitoring does not forward events or authorize cross-account delivery.
C:S3 storage lifecycle configuration does not replace cross-account bus authorization, forwarding, and target error handling.
D:Sending rules still require destination resource authorization; removing it does not solve AccessDenied.
Question 338
A central event bus receives partner events but should not let arbitrary principals inject events. What is the key boundary?
- Restrict event-bus permissions to expected AWS accounts, organizations, and relevant event sources.
- Grant the necessary cross-account event-bus access while retaining narrow publisher scope.
- Configure approved cross-account EventBridge bus policies and sending rules.
- Use EventBridge cross-account rules and resource policies rather than direct per-consumer SDK calls.
Explanation
Correct Answer
A:Resource policies define allowed PutEvents principals; broad trust would weaken audit integrity. Attempt PutEvents from a principal outside the allowed organization.
Incorrect Answers
B:Both sending rules and destination resource policies must authorize event transfer across accounts. Restrict event-bus permissions to expected AWS accounts, organizations, and relevant event sources. Attempt PutEvents from a principal outside the allowed organization.
C:Central event buses accept only authorized publishers and route matched events to audit consumers. Restrict event-bus permissions to expected AWS accounts, organizations, and relevant event sources. Attempt PutEvents from a principal outside the allowed organization.
D:Event bus integration decouples producer implementation from the consumers' account lifecycle. Restrict event-bus permissions to expected AWS accounts, organizations, and relevant event sources. Attempt PutEvents from a principal outside the allowed organization.
Question 339
A security service must consume alerts without granting its application account access to the producer's databases. What should be sent across the account boundary?
- Publish limited, nonsecret event metadata through an authorized event bus.
- Grant the necessary cross-account event-bus access while retaining narrow publisher scope.
- Configure approved cross-account EventBridge bus policies and sending rules.
- Restrict event-bus permissions to expected AWS accounts, organizations, and relevant event sources.
Explanation
Correct Answer
A:A focused event contract communicates incidents without sharing database credentials or broad data access. Inspect the received payload and verify no confidential database fields are included.
Incorrect Answers
B:Both sending rules and destination resource policies must authorize event transfer across accounts. Publish limited, nonsecret event metadata through an authorized event bus. Inspect the received payload and verify no confidential database fields are included.
C:Central event buses accept only authorized publishers and route matched events to audit consumers. Publish limited, nonsecret event metadata through an authorized event bus. Inspect the received payload and verify no confidential database fields are included.
D:Resource policies define allowed PutEvents principals; broad trust would weaken audit integrity. Publish limited, nonsecret event metadata through an authorized event bus. Inspect the received payload and verify no confidential database fields are included.
Question 340
A producer team frequently adds consumers in other AWS accounts; what reduces repeated producer code changes?
- Use EventBridge cross-account rules and resource policies rather than direct per-consumer SDK calls.
- Configure approved cross-account EventBridge bus policies and sending rules.
- Add an EventBridge event pattern on the compliance rule matching the approved classification fields.
- Grant the necessary cross-account event-bus access while retaining narrow publisher scope.
Explanation
Correct Answer
A:Event bus integration decouples producer implementation from the consumers' account lifecycle. Add one consumer rule and verify producer binaries remain unchanged.
Incorrect Answers
B:Central event buses accept only authorized publishers and route matched events to audit consumers. Use EventBridge cross-account rules and resource policies rather than direct per-consumer SDK calls. Add one consumer rule and verify producer binaries remain unchanged.
C:Event-pattern filtering occurs at the bus, avoiding unauthorized or irrelevant consumer invocation. Use EventBridge cross-account rules and resource policies rather than direct per-consumer SDK calls. Add one consumer rule and verify producer binaries remain unchanged.
D:Both sending rules and destination resource policies must authorize event transfer across accounts. Use EventBridge cross-account rules and resource policies rather than direct per-consumer SDK calls. Add one consumer rule and verify producer binaries remain unchanged.