Free Amazon Web Services SAA-C03 Practice Exam with Questions & Answers | Set: 2

HPC workloads require high-throughput, low-latency networking, especially for tightly-coupled applications like weather modeling, genomics, or real-time rendering.

A cluster placement group places instances in the same Availability Zone and on physically connected hardware, reducing network latency and increasing throughput.

Elastic Fabric Adapter (EFA) is a network device for EC2 instances that enables low-latency, high-throughput networking using OS-bypass technology, ideal for tightly-coupled HPC applications.

Each instance can support single-flow 10 Gbps bandwidth using EFA, and collectively, the cluster can achieve up to 100 Gbps aggregate throughput when properly configured.

This solution supports the Performance Efficiency and Resilience design principles and is a standard AWS-recommended pattern for HPC.

???? References:

EC2 Placement Groups

Elastic Fabric Adapter Overview

Best Practices for HPC on AWS

Background Jobs with Event Invocation Type:

The Event invocation type is asynchronous, meaning the Lambda function does not send an immediate result to the API Gateway and processes the request in the background. This is ideal for video encoding tasks that take time.

REST API vs. HTTP API:

REST APIs support advanced features like API keys, request validation, and throttling that HTTP APIs do not support fully.

Since the developer needs API keys and request validations, a REST API is the correct choice.

Integration with Lambda:

AWS Lambda integration is seamless with REST APIs, and using the Event invocation ensures asynchronous processing.

Incorrect Options Analysis:

Option A: HTTP API lacks full support for API keys and validation.

Option CandD: RequestResponse invocation type requires immediate responses, unsuitable for background jobs.

Aurora blue/green deployments are specifically designed for safe schema changes, zero-downtime updates, and production isolation.

The staging (green) environment can receive schema changes without affecting production (blue). After validation, you perform a fast, minimally disruptive switchover that updates production.

Read replicas (Option B) do not allow schema changes. Creating an independent staging cluster (Option A) does not provide automated, low-downtime cutover. DynamoDB (Option D) is not compatible with MySQL schemas.

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

The issue in this case is related to the processing tier, where EC2 instances are overwhelmed at peak times, causing delays.Option D, using anAmazon EC2 Auto Scaling target tracking policybased on theApproximateNumberOfMessagesin the SQS queue, is the best solution.

Auto Scaling with Target Tracking:

Target tracking policiesdynamically scale out or in based on a specific metric. For this use case, you can monitor theApproximateNumberOfMessagesin the SQS queue. When the number of messages (orders) in the queue increases, the Auto Scaling group will scale out more EC2 instances to handle the additional load, ensuring that the queue doesn’t build up and cause delays.

This solution is ideal for handling variable and unpredictable peak times, as Auto Scaling can automatically adjust based on real-time load rather than scheduled times.

Why Not the Other Options?:

Option A (Scheduled Scaling): Scheduled scaling works well for predictable peak times, but this company experiencesunpredictable peak usage, making scheduled scaling less effective.

Option B (ElastiCache for Redis): Adding a caching layer would help if DynamoDB were the bottleneck, but in this case, the issue is CPU overload on EC2 instances in the processing tier.

Option C (CloudFront): CloudFront would help cache static content from the web tier, but it wouldn’t resolve the issue of the processing tier’s overloaded EC2 instances.

AWS References:

Amazon EC2 Auto Scaling Target Tracking

Amazon SQS ApproximateNumberOfMessages

For global low latency, AWS recommends using Amazon CloudFront as a content delivery network in front of both static and dynamic origins when possible.

Static content on Amazon S3 + CloudFront is the standard cost-effective pattern: S3 for low-cost durable storage, CloudFront for global edge caching.

Dynamic content exposed via Amazon API Gateway HTTP API can also be used as an origin for CloudFront, giving global users reduced latency and the ability to cache appropriate responses (when safe).

This pattern minimizes cost by:

Using S3 instead of EC2 for static hosting.

Using HTTP APIs (cheaper than REST APIs) for dynamic content.

Offloading a significant portion of requests to CloudFront cache.

Why the other options are not correct:

B: Static content is not fronted by CloudFront, so global latency is higher.

C: EC2-based web servers for the core web tier add more operational and cost overhead than needed for a mostly static + API pattern.

D: Only the static content uses CloudFront; the dynamic API still lacks a latency-optimized global entry point.

The workload is Windows-based and requires a shared file system with SMB support and very high throughput for 4K video editing.

Amazon FSx for Windows File Server is a fully managed, highly available native Windows file system that supports the SMB protocol, Active Directory integration, and can deliver high throughput and low latency suitable for media workloads.

FSx for Windows File Server supports automatic storage scaling to grow file system capacity as data increases, matching the requirement of +1 TB per week with minimal admin effort.

A Multi-AZ SSD deployment provides high availability and performance.

Why others are not correct:

B: Amazon EFS is NFS, not SMB, and is optimized for Linux clients, not Windows-based environments.

C: FSx for Lustre is a high-performance POSIX file system typically used with Linux-based HPC workloads, not Windows + SMB.

D: S3 File Gateway is not designed for sustained multi-GB/s shared-edit performance and adds additional complexity; it is more suited for backup/archival and limited shared file access.

Amazon S3 is designed for durability, scalability, and millisecond access. For small monthly data volumes (300 MB), S3 Standard is cost-effective and provides immediate access. To meet 30-day retention, Lifecycle policies can automatically expire objects after the required time. For disaster recovery, S3 Cross-Region Replication (CRR) copies objects across Regions to a backup bucket, ensuring data resiliency. OpenSearch (A) is not needed because the requirement is storage and retrieval, not indexing. Aurora or RDS options (C, D) add unnecessary complexity and cost, as a relational database is not required for JSON storage and millisecond retrieval. Therefore, option B provides the simplest, most resilient, and cost-optimized solution.

Amazon S3 Express One Zone provides single-digit millisecond latency and high throughput, making it ideal for ML workloads that require multiple compute nodes and fast access. It is also more cost-effective than traditional file or block storage for temporary, high-speed needs.

Usingcross-account IAM rolesis the most secure and scalable way to share data between AWS accounts.

Atrust relationshipallows the analytics team's account to assume the role in the main account and access the DynamoDB table directly.

Ais feasible but involves data duplication and additional costs for storing the JSON files in S3.

B and Cviolate security best practices by allowing public access to sensitive data and sharing credentials, which is highly discouraged.

AWS Documentation Reference:

Cross-Account Access with Roles

Best Practices for Amazon DynamoDB Security

For high availability, Amazon ElastiCache for Redis supports Multi-AZ with automatic failover, which provides primary and replica nodes in different Availability Zones. If the primary node fails, Redis automatically promotes a replica to primary.

From AWS Documentation:

“When you enable Multi-AZ with automatic failover, Amazon ElastiCache automatically detects failures and promotes a read replica to primary with minimal downtime.”

(Source: Amazon ElastiCache for Redis User Guide)

Why B is correct:

Multi-AZ provides automatic failover and data replication.

Ensures continuous availability and protects against node or AZ failures.

Fully managed with no manual intervention needed.

Why others are incorrect:

A: Manual promotion is not automatic.

C & D: Restoring from backup is slow and meant for disaster recovery, not failover.

Amazon EFS is a “fully managed, elastic, shared file system” that provides NFSv4 access across many EC2 instances and AZs, ideal for lift-and-shift of Linux NFS workloads. AWS DataSync “automates moving data between on-premises NFS servers and Amazon EFS,” performing incremental, parallel transfers with encryption and integrity checks, enabling cutover with minimal disruption. FSx for Lustre (A) targets HPC and S3-backed workloads, not general NFS shares. FSx for Windows (D) exposes SMB, not NFS. S3 (C) is object storage and not directly NFS-accessible by multiple EC2 hosts. EFS + DataSync satisfies NFS compatibility, multi-client access, and low-touch migration while remaining cost-effective for a 200-GB dataset.

IAM Role: Attaching an IAM role to an EC2 instance profile is a secure way to manage permissions without embedding credentials.

AWS Secrets Manager: Grants controlled access to database credentials and automatically rotates secrets if configured.

Identity-Based Policy: Ensures the IAM role only has access to specific secrets, enhancing security.

AWS Secrets Manager Documentation

AWS documentation states that the correct and least-privilege method for cross-account SNS-to-SQS integration is:

• Add specific SNS topic ARNs to the SQS queue policy.

• Allow only those topics to publish messages to the queue.

• Ensure SNS has permission to publish to the specific queue ARN.

This ensures strict scoping and adheres to least privilege.

Options A and D grant overly broad permissions. Option B allows publishing to any queue, which violates least privilege.

DynamoDB Accelerator (DAX) is a fully managed, in-memory cache for DynamoDB that delivers up to a 10x performance improvement—even at millions of requests per second. DAX requires minimal changes to existing applications and offloads the read workload from DynamoDB during report generation.

Reference Extract:

"DAX provides in-memory acceleration for DynamoDB tables, reducing response times from milliseconds to microseconds with minimal application changes."

Source: AWS Certified Solutions Architect – Official Study Guide, DynamoDB and Performance section.

AWS CloudFormationallows for the automated, repeatable setup of infrastructure, reducing human error and ensuring consistency.AWS Configprovides the ability to track changes in the infrastructure, ensuring that all changes are logged and auditable, which satisfies the requirement for tracking incremental changes.

Option A and C (AWS Organizations): AWS Organizations manage multiple accounts, but they are not designed for infrastructure setup or change tracking.

Option D (Service Catalog): Service Catalog is used for deploying products, not for setting up infrastructure or tracking changes.

AWS References:

AWS Config

AWS CloudFormation