Amazon AWS Certified Solutions Architect - Associate SAA-C03 Practice Test 26

Question 250

A Lambda API function opens connections to an RDS database with limited session capacity. Sudden bursts of invocations overwhelm the DB. Which Lambda control can directly cap concurrent handlers?

Set the function memory as high as possible and expect fewer concurrent invocations automatically.
Increase API Gateway cache TTL and assume all writes become safe.
Increase the RDS max_connections setting and rely on Lambda burst concurrency to fall automatically as database session pressure rises.
Configure reserved concurrency for the function based on downstream capacity and queue/throttle excess requests appropriately.

Explanation

Correct Answer

Reserved concurrency provides a per-function limit that helps prevent Lambda bursts from opening more database work than it can support.

Incorrect Answers

Memory allocation affects resources per invocation, not an upper concurrency limit.

Caching does not guarantee a hard concurrency limit on all operations.

Raising the database limit only delays saturation; it does not place a hard bound on Lambda simultaneous execution or shield the database from uncontrolled bursts.

Question 251

A Lambda SQS batch contains 10 records; one message has a transient error while nine succeeded. The app must avoid unnecessarily retrying every successful record. Which integration behavior should it use?

Use supported partial batch response reporting so failed message identifiers are retried while successfully processed records can be acknowledged.
Return one generic function error and assume Lambda deletes all ten messages.
Delete every SQS message before processing to avoid retries.
Enable a longer SQS visibility timeout but return an unhandled Lambda exception for one failed record, expecting Lambda to acknowledge the other nine automatically.

Explanation

Correct Answer

SQS-Lambda partial batch failure reporting avoids whole-batch retries when only part of a batch fails, with proper idempotent consumers.

Incorrect Answers

A failed batch can cause retried records and duplicate work.

Early acknowledgement risks losing unfinished work.

A longer visibility period delays redelivery; without ReportBatchItemFailures and a valid partial-batch response, a failed invocation can cause previously successful messages to be retried.

Question 252

A serverless REST API needs authenticated HTTPS routing, method-level throttling and integration with multiple Lambda handlers. Which managed front door is designed for this?

Use Amazon API Gateway with routes, authorizers and quotas/throttles integrated with the Lambda functions.
Place a public IP on every Lambda function instance.
Use SQS as a direct synchronous HTTP path router for arbitrary browsers.
Use an ALB HTTPS listener with Lambda targets and assume it provides every API Gateway authorizer, per-client usage plan and token quota without additional components.

Explanation

Correct Answer

API Gateway manages public API ingress and controls without having to run a reverse-proxy EC2 fleet.

Incorrect Answers

Lambda execution environments do not expose application-owned public IP instances for direct HTTP serving.

SQS is asynchronous messaging, not an API protocol gateway.

ALB can route HTTP and invoke Lambda targets but does not automatically provide all API Gateway REST authorizer/usage-plan capabilities and detailed API lifecycle controls.

Question 253

A SaaS hosts catalog and checkout services in separate EC2 target groups. Both use the same HTTPS hostname but /catalog and /checkout must scale independently. Which routing component fits?

Use a Network Load Balancer that parses URL paths as its native L7 routing condition.
Give both services one hard-coded EC2 IP and ignore separate target health.
Use Route 53 weighted routing between two ALB hostnames and expect DNS to inspect /catalog versus /checkout on each individual HTTPS request.
Configure Application Load Balancer path-based listener rules routing each prefix to its respective target group.

Explanation

Correct Answer

ALB examines HTTP request attributes and forwards to separately managed and scaled target groups.

Incorrect Answers

NLB is transport-oriented and does not natively provide ALB-style HTTP path rules.

One fixed endpoint eliminates independent routing and resilience.

DNS can select hosts but does not parse URL paths; ALB listener path conditions direct different URL prefixes to distinct target groups.

Question 254

A small engineering team runs long-lived OCI containers but does not want to patch or scale EC2 worker nodes. Which managed ECS compute option is appropriate?

Run EC2 self-managed nodes because Fargate requires manually maintained kernel patches.
Use Lambda as the guaranteed host for an indefinitely long-running container server.
Deploy an EKS managed node group and assume Kubernetes worker nodes require no operating-system or node lifecycle management from the customer.
Use Amazon ECS services/tasks on AWS Fargate with appropriate task definitions, networking and autoscaling.

Explanation

Correct Answer

Fargate supplies managed container compute capacity without managing the underlying EC2 instance fleet.

Incorrect Answers

Fargate abstracts the host management responsibilities the team wants to avoid.

Lambda invocation/runtime limits differ from long-lived server containers.

EKS manages the control plane but a managed node group still has instance lifecycle and upgrade responsibilities; ECS on Fargate avoids provisioning and patching an EC2 worker fleet.

Question 255

An enterprise has existing Kubernetes manifests, controllers and admission policies. It wants a managed Kubernetes control plane while retaining Kubernetes API integration. Which AWS compute platform best fits?

Use an ECS-only API and assume all Kubernetes controllers run unchanged.
Use S3 event notifications as the managed Kubernetes control plane.
Run the same Kubernetes manifests on a standalone ECS cluster without translating Kubernetes APIs, admission controllers or custom resources.
Use Amazon EKS with a supported node/Fargate strategy and operational responsibility for compatible cluster components.

Explanation

Correct Answer

EKS offers managed Kubernetes control plane while preserving the upstream Kubernetes orchestration model and app contracts.

Incorrect Answers

ECS is a distinct scheduler and does not expose identical Kubernetes APIs.

Object events cannot run Kubernetes operators/controllers.

ECS is a different container orchestrator and does not implement the complete Kubernetes API and controller architecture that these workloads require.

Question 256

A stateless web tier sees highly variable request volume and must replace failed instances automatically. Which AWS mechanism should keep healthy capacity aligned with load?

Use CloudWatch alarms and SNS notifications to tell an operator when CPU rises, assuming receiving an alert creates and registers healthy EC2 instances automatically.
Use a static single EC2 instance and rely on CloudWatch monitoring to add capacity automatically.
Use DynamoDB capacity scaling in place of EC2 web server scaling.
Use an EC2 Auto Scaling group across suitable AZs with health checks and a scaling policy based on measured demand.

Explanation

Correct Answer

Auto Scaling adjusts desired capacity and replaces unhealthy instances while preserving a fleet rather than one manually managed host.

Incorrect Answers

Observability can report need for capacity but does not by itself provision a managed fleet or replace unhealthy hosts. Auto Scaling groups act on scaling policies and health checks.

Monitoring alone does not provision replacement instances.

Database capacity cannot replace the stateless web fleet.

Question 257

An HTTP application stores user sessions only in local EC2 memory. When its ALB routes the next request to a different instance, the user appears logged out. Which architecture supports horizontal scaling?

Move session state into a shared appropriately secured store such as ElastiCache, with stateless application instances.
Require permanent ALB sticky sessions and assume instance failures preserve local memory.
Disable instance health checks to prevent session migration.
Enable an ALB stickiness cookie as the sole state-management system and assume the cookie preserves an EC2 host’s lost in-memory session data after replacement.

Explanation

Correct Answer

Centralized session data lets any healthy backend instance continue the user session independently of original instance placement.

Incorrect Answers

Stickiness can limit distribution but cannot recover in-memory state after instance termination.

Unhealthy instances would remain in service and state loss is not addressed.

Stickiness can help route a user to one available host, but it does not make its volatile memory durable across host replacement. A shared session store removes that failure dependency.

Question 258

A burst of Lambda invocations connects to a small RDS database and exhausts its connection limit. The queries themselves are quick. Which managed feature can pool/reuse connections?

Configure additional read replicas but leave every Lambda invocation creating new short-lived write connections directly to the same primary writer.
Use Amazon RDS Proxy with properly scoped authentication and connection pooling for the supported database.
Increase S3 transfer acceleration throughput.
Make the database publicly accessible so it accepts more sessions.

Explanation

Correct Answer

RDS Proxy pools database connections for many application clients, lowering connection churn and reducing concurrent backend sessions.

Incorrect Answers

Readers may offload suitable queries but do not pool the writer connections that this scenario exhausts. RDS Proxy provides managed database connection reuse.

Object transfer optimization is unrelated to database session pooling.

Network exposure does not increase configured DB connection capacity.

Question 259

A global news application serves rarely changing images and JavaScript from an S3 origin. Readers far from the origin see high latency and repetitive object reads. Which AWS service improves delivery?

Configure Route 53 latency-based DNS to direct all browsers to the single existing S3 bucket Region without caching objects near consumers.
Use Amazon CloudFront edge caching with appropriate cache policies and origin protections.
Use SQS FIFO to synchronously serve web images to browser clients.
Disable HTTP caching everywhere and require every client to reach the origin.

Explanation

Correct Answer

CloudFront serves cacheable content from nearby edges and reduces repeated origin requests, improving latency and backend load.

Incorrect Answers

DNS location routing alone does not create edge copies of S3 objects or reduce repeated origin GETs. CloudFront caches content near viewers using defined cache rules.

A queue is not an edge cache or HTTP file-serving network.

That increases distance and repeated origin reads.

Question 260

A public REST API fronts a third-party service limited to 100 requests per second. Clients may burst well above that. Which gateway behavior prevents overload?

Set all requests to unlimited concurrency and rely only on the third party returning 429.
Set a CloudFront response cache TTL for API GET requests and assume this imposes a strict global 100-call-per-second ceiling on all uncached API methods.
Configure API Gateway token-bucket throttling/usage plans as protective targets, and add a downstream concurrency limiter or queue if the third-party 100-RPS ceiling is strict.
Increase the S3 object retention period for the API host.

Explanation

Correct Answer

API Gateway throttles and quotas lower the flow of bursts to an upstream dependency, but AWS documents those limits as best-effort targets, not guaranteed hard ceilings. Protect a truly strict contract with an application-side token bucket, bounded integration concurrency or durable queue and careful backoff.

Incorrect Answers

This overloads the dependency and increases failures before protective controls act.

Caching may reduce origin requests for eligible responses, but it does not impose a reliable requests-per-second quota on all uncached API operations. Use throttling plus downstream protection.

Storage retention cannot limit backend request throughput.

Pass Your Amazon AWS Certified Solutions Architect - Associate SAA-C03 Exam Easy!

Question 250

Explanation

Correct Answer

Incorrect Answers

Question 251

Explanation

Correct Answer

Incorrect Answers

Question 252

Explanation

Correct Answer

Incorrect Answers

Question 253

Explanation

Correct Answer

Incorrect Answers

Question 254

Explanation

Correct Answer

Incorrect Answers

Question 255

Explanation

Correct Answer

Incorrect Answers

Question 256

Explanation

Correct Answer

Incorrect Answers

Question 257

Explanation

Correct Answer

Incorrect Answers

Question 258

Explanation

Correct Answer

Incorrect Answers

Question 259

Explanation

Correct Answer

Incorrect Answers

Question 260

Explanation

Correct Answer

Incorrect Answers

AWS Certified Solutions Architect - Associate SAA-C03 Practice Tests