Amazon Web Services SAA-C03 - AWS Certified Solutions Architect - Associate (SAA-C03)

Enabling Multi-AZ deployment for an Amazon RDS DB instance is the simplest and most effective way to improve database availability and eliminate a single point of failure. Multi-AZ deployments provide automatic failover to a standby instance in case of a failure of the primary instance. This approach is managed by AWS and only requires a modification to the existing instance, with minimal manual intervention or downtime, thus providing the least implementation effort.

Reference Extract from AWS Documentation / Study Guide:

" With Multi-AZ deployments, Amazon RDS automatically creates a primary DB Instance and synchronously replicates the data to a standby instance in a different Availability Zone (AZ). In case of a failure of the primary DB Instance, Amazon RDS automatically fails over to the standby so that database operations can resume quickly. "

Source: AWS Certified Solutions Architect – Official Study Guide, RDS High Availability section.

Aservice control policyis the correct organization-wide control because SCPs define the maximum available permissions across member accounts. AWS documentation also notes that in SCPs you can use the global condition keyaws:PrincipalArnin theConditionelement to identify specific principals, including roles. That makes option D the correct pattern: deny modification of Lambda functions for everyone except the designated security-team role. An SCP that denies all changes with no exception would block the security team too. The Principal element is not the right way to express this exception in an SCP for this use case. Therefore, the correct answer is an SCP with an exception in the Condition clause for the security team role.

The critical requirement is immediate processing upon customer submission, with a typical compute duration of about 10 seconds. This is an ideal fit for a synchronous, request-driven, serverless pattern: API Gateway + AWS Lambda. API Gateway provides a managed HTTPS front door for the application, handles request validation/throttling if needed, and integrates directly with Lambda. Lambda is designed for short-lived computations, scales automatically with request volume, and can complete 10-second tasks well within typical Lambda execution limits.

Option A therefore meets the requirement with low latency and minimal operational overhead. When a customer submits data, API Gateway invokes the Lambda function immediately, and the function returns the calculated premium response. This preserves an interactive user experience and avoids provisioning or managing servers. It also handles bursts gracefully because Lambda concurrency can scale quickly, and API Gateway can absorb high request rates.

Option B is unsuitable because launching a Spot instance per upload introduces significant startup latency (instance launch time) and is not reliable due to Spot interruptions. Option C and D are not “immediate”: Transfer Family is for file transfers, and scheduled EKS jobs or Batch jobs introduce queuing and scheduling delays. While Batch can be event-driven, it is typically used for longer-running or batch-oriented workloads and still has job-start overhead compared to Lambda.

Therefore, A best matches the requirements for immediate execution, scalability, and low operational burden while maintaining responsive performance for customers.

By exporting AWS Cost and Usage Reports (CUR) to Amazon S3 and analyzing them with Amazon QuickSight, companies can generate interactive visual dashboards. This solution is fully AWS-managed, requires no third-party tools, and integrates deeply with AWS cost data.

BothAmazon KinesisandSQS FIFOqueues ensure the sequential processing of messages. By using the payment ID as the partition key in Kinesis or as the message group in the SQS FIFOqueue, messages are processed in order. Both solutions also allow for long-term retention (up to 10 days) of messages, making them suitable for this payment processing use case.

Option A (DynamoDB): DynamoDB does not guarantee message ordering for real-time processing.

Option C (ElastiCache): ElastiCache is for caching, not suitable for sequential message processing.

Option D (Standard SQS queue): A standard SQS queue does not guarantee ordering of messages.

AWS References:

Amazon Kinesis

Amazon SQS FIFO Queues

The primary requirements are automatic scaling, resilience to traffic spikes, and guaranteed order durability without losing customer orders. Option D follows a well-established, AWS-recommended decoupled architecture pattern that meets all these requirements.

Using an Application Load Balancer (ALB) provides Layer 7 request routing and distributes incoming traffic across multiple EC2 instances in Auto Scaling groups, allowing the application layer to scale horizontally as traffic increases. This ensures that spikes in web traffic do not overwhelm individual instances.

The critical resilience feature in this design is Amazon SQS, which acts as a durable buffer for unprocessed customer orders. If the application experiences a sudden traffic surge or if downstream processing slows, SQS stores incoming orders reliably until they can be processed. This decoupling prevents request loss and protects the user experience by absorbing load bursts.

Processed orders are stored in Amazon RDS with a Multi-AZ deployment, which provides synchronous replication to a standby instance in another Availability Zone and automatic failover. This ensures high availability and data durability for transactional order data.

The other options misuse services or introduce unnecessary complexity. NAT gateways (Option A) do not manage web traffic or provide resiliency for application workloads. Redshift (Option B) is an analytics data warehouse and not suitable for transactional order processing. Gateway Load Balancer (Option C) is designed for network appliances, not application traffic management or order capture.

Therefore, D is the correct solution because it combines load balancing, auto scaling, message queuing, and highly available databases to deliver a resilient, scalable, and fault-tolerant order processing architecture.

Durable storage of incoming orders with buffering and ability to handle surges is exactly what Amazon SQS is designed for. SQS provides highly durable, scalable queues that decouple producers from consumers.

Centrally tracking workflow steps is a core use case of AWS Step Functions, which gives a visual workflow and state machine, tracks the state of each order, and can orchestrate calls to multiple microservices (in this case, Lambda functions).

Combining SQS + Step Functions + Lambda gives:

Durable queueing for orders (SQS).

Loose coupling and surge handling (SQS decoupling + auto-scaling Lambda).

Central orchestration and tracking of order-processing steps (Step Functions).

Why the other options are not correct:

A: SNS is a pub/sub service, not a durable work queue, and is not designed for “store-and-retry until processed” workloads in the same way SQS is.

C: SQS + EventBridge provides decoupling but no central, stateful workflow tracking; EventBridge is event routing, not workflow orchestration.

D: SNS + EventBridge still lacks durable order storage and explicit centralized workflow/state tracking.

S3 Block Public Access only blocks public access—not explicitly allowed access.

An S3 bucket policy can explicitly permit access only from specific source IP addresses using the aws:SourceIp condition. This allows secure, direct HTTP access to the S3 object from known firewall IP addresses.

Static website hosting (Option A) requires the bucket to be public and is blocked by the enabled S3 Block Public Access setting.

CloudFront with OAC (Option C) is unnecessary and adds cost and complexity.

Lambda (Option D) introduces operational overhead and is not needed since S3 policies can enforce IP restrictions directly.

The requirement is for workloads inside a VPC to resolve private DNS names that are hosted on-premises over a VPN connection. The most secure and AWS-recommended method to achieve this is to use Amazon Route 53 Resolver outbound endpoints with forwarding rules.

Option A enables DNS queries that originate in the VPC to be securely forwarded to on-premises DNS servers through the VPN connection. A Route 53 Resolver outbound endpoint provides elastic, managed DNS forwarding capability without exposing DNS services to the public internet. By creating a Resolver rule, the company can define which domain names (for example, internal corporate domains) should be forwarded to specific on-premises DNS IP addresses. Associating the rule with the VPC ensures only authorized VPCs can use this resolution path.

Option B is used for the opposite scenario: when on-premises systems need to resolve private DNS names hosted in AWS, not when AWS needs to resolve on-premises names. Option C is invalid because Route 53 private hosted zones can only be associated with VPCs, not directly with on-premises networks. Option D is insecure because it exposes internal service records publicly and violates the requirement for private DNS.

Therefore, A is the most secure solution because it keeps DNS resolution private, uses managed AWS infrastructure, integrates cleanly with VPN connectivity, and follows AWS hybrid DNS best practices.

Using an Application Load Balancer (ALB) allows for integration with AWS WAF to inspect all incoming traffic. Running the application on Amazon ECS provides a scalable and managed container orchestration service. Utilizing Amazon Aurora Serverless for the PostgreSQL database offers automatic scaling based on application demand, which is cost-effective for workloads with variable traffic patterns, such as higher traffic on weekdays and lower traffic on weekends.

Amazon Kinesis Data Streams provides a highly scalable and durable service for ingesting real-time streaming data. By decoupling ingestion and processing, Kinesis can handle large spikes in traffic without service disruption. Lambda functions (or other consumers) can then process the data as it arrives, scaling automatically. This pattern avoids 503 errors due to compute saturation and delivers a resilient, serverless, and highly scalable architecture.

Reference Extract from AWS Documentation / Study Guide:

" Kinesis Data Streams provides a scalable and durable real-time data streaming service. Coupling Kinesis with AWS Lambda enables event-driven processing, elasticity, and decoupling between ingestion and processing layers. "

Source: AWS Certified Solutions Architect – Official Study Guide, Streaming and Serverless section.

AWS WAF is the correct service because AWS documents that it protects web applications at Layer 7 and defends against common web exploits such as SQL injection and cross-site scripting. Associating WAF with the ALB allows malicious requests to be filtered before they reach the EC2 instances, which minimizes application impact and avoids pushing inspection logic onto the hosts. CloudTrail, GuardDuty, and network ACLs are not the best tools for active application-layer request filtering against SQLi and XSS. Shield Advanced can add further protection for broader attack scenarios, but the key control in this answer is WAF on the ALB with rules that block malicious traffic patterns.

============

Amazon Timestreamis purpose-built for storing and analyzing time-series data like telemetry.

Option Aleverages Timestream, SageMaker for ML, and QuickSight for visualization, meeting all requirements with minimal complexity.

Option Binvolves more complex DynamoDB-EMR integration.

Option Cuses Neptune, which is designed for graph databases, not telemetry data.

Option Dincorrectly uses Athena for visualization instead of QuickSight.

The best answer ismessage batching. AWS documentation explains that SQS batch actions let applications send, receive, or delete up to 10 messages at a time. AWS also notes that batching can be combined with horizontal scaling to improve throughput with fewer requests and connections. That makes batching the most relevant feature for handling sudden traffic increases efficiently. FIFO queues address strict ordering and deduplication, visibility timeout controls how long a received message stays hidden from other consumers, and long polling reduces empty responses and unnecessary polling cost. None of those features directly improve throughput as effectively as batching for traffic spikes.

============

Auto Scalingis the most effective solution for ensuring seamless scalability of a stateless application. Key points:

Dleverages Auto Scaling with a Spot Fleet for cost efficiency and attaches an ALB to distribute traffic.

A and Bdo not provide automated scaling and would require manual intervention to add more instances.

Cchanges the instance type but does not scale out horizontally, which is required here.

AWS Documentation Reference:

Amazon EC2 Auto Scaling

For globally distributed consumers of REST APIs, API Gateway edge-optimized endpoints “route requests to the nearest CloudFront edge location, which then forwards them to your API in the [home] Region,” reducing latency for users worldwide and scaling automatically for unknown or spiky traffic. By contrast, Regional endpoints are intended for clients within the same or nearby Region and do not provide global edge acceleration. AWS Lambda provides automatic scaling for serverless backends; latency can be further reduced by right-sizing memory (which also increases CPU) and, when needed, enabling provisioned concurrency to keep functions initialized and eliminate cold starts for critical paths. Using NLBs across Regions is not serverless and adds operational complexity without CloudFront edge acceleration. Therefore, combining API Gateway edge-optimized REST APIs with Lambda meets the requirements of minimal global latency and unknown initial traffic.

For Amazon RDS relational databases experiencing read-heavy workloads, AWS best practice is to create read replicas and offload read traffic to those replicas. A read replica:

Uses asynchronous replication from the primary.

Can serve read-only queries, thereby reducing load and query latency on the primary.

Can be used behind an application-level read/write splitting mechanism or via endpoints that direct read traffic to replicas.

Performance Insights (Option A) helps diagnose performance problems but does not itself offload traffic.

DAX (Option B) is a cache for DynamoDB, not RDS.

Kinesis Data Firehose (Option D) is a streaming ingestion service and cannot increase RDS query concurrency.

When AWS workloads must resolve DNS names from on-premises systems over a hybrid network (like VPN or Direct Connect), the best solution is to use Amazon Route 53 Resolver outbound endpoints.

The outbound endpoint enables DNS queries to be forwarded from your VPC to on-premises DNS servers.

You must also configure a Route 53 Resolver forwarding rule to define which domain names (e.g., corp.internal) should be forwarded to the specific on-premises DNS IPs.

This setup allows private DNS resolution from AWS to on-premises systems and is fully managed, eliminating the need to run and maintain EC2-based DNS proxies (as in option A).

Options C and D are incorrect:

C is not DNS-specific and doesn’t solve name resolution.

D misuses a public hosted zone for a private DNS domain.

🔗 Reference:

Route 53 Resolver Outbound Endpoints

Amazon S3 Transfer Acceleration is designed specifically to speed up long-distance uploads and downloads to an S3 bucket by using globally distributed CloudFront edge locations and optimized AWS network paths. Because users are located worldwide and upload images to one centralized S3 bucket, Transfer Acceleration directly addresses the performance bottleneck. The important detail is that clients must use the S3 Accelerate endpoint; simply enabling Transfer Acceleration but continuing to use the standard endpoint would not provide the acceleration benefit. CloudFront is useful for caching and accelerating content delivery to users, but it is not the best answer for improving global upload speed into a centralized S3 bucket.