IT Cert Ready

AWS

AWS Certified Solutions Architect – Associate

Exam code: SAA-C03

Start practicing

About this certification

AWS Certified Solutions Architect – Associate is a AWS certification. Exam code SAA-C03. IT Cert Ready offers 1,188 practice questions with domain quizzes, full mock exams, and detailed explanations. Practice by official domain with 4 blueprint-aligned domains.

Official exam domain weights

Verified blueprint weights from the official exam outline.

  • Domain 1 · Design Secure Architectures · 30%
  • Domain 2 · Design Resilient Architectures · 26%
  • Domain 3 · Design High-Performing Architectures · 24%
  • Domain 4 · Design Cost-Optimized Architectures · 20%

Sample AWS Certified Solutions Architect – Associate practice questions

Preview realistic multiple-choice items from the IT Cert Ready question bank. Expand each answer for the correct choice and explanation—the same format used in domain quizzes and mock exams.

  1. Sample question 1 · Design Cost-Optimized Architectures (20%)

    Patchline Systems is a SaaS startup headquartered in Germany that provides automated patch-cycle orchestration to regulated financial services clients across the EU. The company stores patch compliance records, vulnerability scores, and audit trails in Amazon Aurora MySQL. The database is provisioned as a two-node Aurora cluster (writer + reader) running continuously. The CFO has raised a concern in the quarterly governance review: the Aurora cluster accounts for 38% of total AWS spend, yet telemetry shows the database receives zero queries between 22:00 and 06:00 CET every night and is lightly loaded on weekends. The Head of Engineering argues that the cluster must remain available for on-call engineers who may need ad-hoc queries at any hour. The solutions architect must present a cost-optimized design to the steering committee that balances availability for on-call access with elimination of unnecessary idle spend, without changing the MySQL wire protocol used by the application. As the solutions architect presenting to the Patchline Systems steering committee, which design best balances cost optimization during idle hours with on-call availability and MySQL protocol compatibility?

    • A. Convert the Aurora MySQL cluster to Aurora Serverless v2 with a minimum ACU of 0.5 and enable the cluster to scale to zero during defined idle windows, while retaining the MySQL-compatible cluster endpoint for on-call engineers and the application.
    • B. Delete the Aurora reader node to eliminate its cost and rely solely on the writer node, scheduling a Lambda function to create a new reader node each morning using the AWS SDK.
    • C. Migrate the Aurora MySQL cluster to Amazon RDS for MySQL with Multi-AZ enabled and purchase a 3-year Reserved Instance to reduce the hourly rate.
    • D. Enable Aurora Auto Scaling on the reader node to scale reader replicas to zero during idle hours, while keeping the writer node provisioned at full capacity around the clock.
    View answer & explanation

    Correct answer: A. Convert the Aurora MySQL cluster to Aurora Serverless v2 with a minimum ACU of 0.5 and enable the cluster to scale to zero during defined idle windows, while retaining the MySQL-compatible cluster endpoint for on-call engineers and the application.

    [Domain 4 - Design Cost-Optimized Architectures, Subtopic 4 - Design Cost-Optimized Architectures - Area 1 - Serverless DB] Core Concept: Design cost-optimized database solutions. requires applying serverless db in the context of the scenario, consistent with the official (ISC)² exam outline. Correct Answer (A): Convert the Aurora MySQL cluster to Aurora Serverless v2 with a minimum ACU of 0.5 and enable the cluster to scale to zero during defined idle windows, while retaining the MySQL-compatible cluster endpoint for on-call engineers and the application. Why the other options are incorrect: Option B: Dynamically deleting and recreating a reader node via Lambda introduces operational complexity, potential data consistency risks, and cold-start delays that are unacceptable for on-call engineers who need immediate ad-hoc query access; it also does not address the writer node's idle cost. Option C: Migrating to RDS for MySQL with Multi-AZ and a Reserved Instance locks Patchline into a fixed provisioned cost for three years and does not reduce idle spend — it simply lowers the per-hour rate, which is a pricing optimization rather than an architectural one that addresses the 38% spend concern. Option D: Aurora Auto Scaling manages the number of reader replicas based on load metrics but cannot scale the writer node; the writer node — which is the primary cost driver in a two-node cluster — remains fully provisioned and billed at full capacity throughout the idle window.

  2. Sample question 2 · Design Cost-Optimized Architectures (20%)

    Crestline Energy Systems, a large investor-owned utility serving 3.2 million customers across the Midwest, has deployed a cloud-native network intrusion detection platform on AWS. The platform ingests NetFlow records from 600 transmission sites and runs continuous anomaly detection using a microservices architecture. Individual detection microservices are event-driven, execute in under 900 milliseconds on average, and are triggered by Amazon Kinesis Data Streams as new flow records arrive. Current compute costs are $31,000 per month using a fleet of always-on t3.medium EC2 instances that are significantly underutilized during off-peak hours (average CPU utilization: 11%). The security architecture team reports monthly on cost per detection event, idle compute ratio, and total detection throughput. The team wants to eliminate idle compute costs and improve the cost-per-detection-event metric without adding operational overhead for server management. Which architectural change best meets these goals? As the solutions architect, which compute redesign best eliminates idle capacity costs and improves the cost-per-detection-event KPI for Crestline Energy Systems' event-driven intrusion detection microservices?

    • A. Replace the t3.medium fleet with Spot Instances using a mixed instance policy, and configure an Auto Scaling group to scale in during off-peak hours based on Kinesis stream iterator age.
    • B. Rightsize the t3.medium instances to t3.nano to reduce per-instance cost, and use AWS Compute Optimizer recommendations monthly to track rightsizing opportunities.
    • C. Migrate the detection microservices to AWS Lambda functions triggered directly by Amazon Kinesis Data Streams, and publish cost-per-invocation and total invocation count metrics to Amazon CloudWatch for monthly KPI reporting.
    • D. Purchase a 3-year Standard Reserved Instance commitment for t3.medium across all Availability Zones to lock in the maximum discount and reduce monthly spend.
    View answer & explanation

    Correct answer: C. Migrate the detection microservices to AWS Lambda functions triggered directly by Amazon Kinesis Data Streams, and publish cost-per-invocation and total invocation count metrics to Amazon CloudWatch for monthly KPI reporting.

    [Domain 4 - Design Cost-Optimized Architectures, Subtopic 4 - Design Cost-Optimized Architectures - Area 1 - Spot] Core Concept: Design cost-optimized compute solutions. requires applying spot in the context of the scenario, consistent with the official (ISC)² exam outline. Correct Answer (C): Migrate the detection microservices to AWS Lambda functions triggered directly by Amazon Kinesis Data Streams, and publish cost-per-invocation and total invocation count metrics to Amazon CloudWatch for monthly KPI reporting. Why the other options are incorrect: Option A: Spot Instances with Auto Scaling reduce idle costs but do not eliminate them — the fleet still maintains a minimum capacity baseline, and Spot interruptions introduce risk for a real-time security detection workload; server management overhead also remains. Option B: Rightsizing to t3.nano reduces per-instance cost marginally but does not address the fundamental problem of idle capacity during off-peak hours; at 11% average CPU utilization, the workload would likely be CPU-constrained on nano instances during peak ingestion bursts. Option D: A 3-year Reserved Instance commitment locks in a discount on always-on capacity but does not eliminate idle compute costs; with only 11% average utilization, the utility would pay for 89% unused capacity for three years, worsening the idle compute ratio KPI rather than improving it.

  3. Sample question 3 · Design Cost-Optimized Architectures (20%)

    Vaultrix Security, a SaaS startup offering cloud-based vulnerability management to mid-market enterprises, operates a hybrid cloud environment spanning AWS and an on-premises data center. The engineering team maintains a PostgreSQL database on Amazon RDS to store scan results, remediation tickets, and patch cycle metadata. The CTO has flagged that the RDS instance runs at less than 10% CPU utilization on weekdays and is nearly idle on weekends, yet the company pays for a continuously provisioned db.r6g.2xlarge instance. The board has asked the solutions architect to present a cost-optimization proposal that maintains full SQL compatibility, eliminates over-provisioning costs, and requires minimal changes to the existing application connection strings. The governance team requires that any proposed change be documented with projected cost impact before implementation. As the solutions architect at Vaultrix Security, which database design change best addresses the over-provisioning cost problem while satisfying the governance team's requirement for minimal application refactoring and full PostgreSQL compatibility?

    • A. Migrate the RDS PostgreSQL instance to Amazon Aurora Serverless v2 with a minimum ACU setting of 0.5 and a maximum ACU aligned to peak workload, retaining the same PostgreSQL-compatible endpoint so application connection strings require no change.
    • B. Replace the RDS PostgreSQL instance with Amazon DynamoDB, redesign the schema as key-value pairs, and use DynamoDB on-demand capacity mode to eliminate idle costs.
    • C. Downsize the existing RDS PostgreSQL instance to a db.t4g.medium Reserved Instance and accept the risk of performance degradation during peak scan windows.
    • D. Deploy an Amazon ElastiCache for Redis cluster in front of the existing RDS instance to cache scan results and reduce read load, keeping the db.r6g.2xlarge instance unchanged.
    View answer & explanation

    Correct answer: A. Migrate the RDS PostgreSQL instance to Amazon Aurora Serverless v2 with a minimum ACU setting of 0.5 and a maximum ACU aligned to peak workload, retaining the same PostgreSQL-compatible endpoint so application connection strings require no change.

    [Domain 4 - Design Cost-Optimized Architectures, Subtopic 4 - Design Cost-Optimized Architectures - Area 1 - Serverless DB] Core Concept: Design cost-optimized database solutions. requires applying serverless db in the context of the scenario, consistent with the official (ISC)² exam outline. Correct Answer (A): Migrate the RDS PostgreSQL instance to Amazon Aurora Serverless v2 with a minimum ACU setting of 0.5 and a maximum ACU aligned to peak workload, retaining the same PostgreSQL-compatible endpoint so application connection strings require no change. Why the other options are incorrect: Option B: DynamoDB is a NoSQL key-value and document store that is not PostgreSQL-compatible; migrating to it would require a full schema redesign and significant application refactoring, violating the minimal-change requirement and introducing substantial engineering risk. Option C: Downsizing to a Reserved Instance reduces the hourly rate but does not eliminate idle costs during low-utilization periods; the instance still runs continuously, and accepting performance degradation during peak scan windows is an unacceptable trade-off that the governance team would not approve without a formal risk acceptance process. Option D: Adding ElastiCache reduces read pressure on the RDS instance but does not address the root cause — the instance remains fully provisioned and billed at the db.r6g.2xlarge rate even when idle, so the over-provisioning cost problem is not resolved.

  4. Sample question 4 · Design Cost-Optimized Architectures (20%)

    Vanthorpe Precision Components, a mid-sized manufacturing firm in the Midwest, operates an IoT telemetry platform that collects sensor readings from hundreds of CNC machines on the factory floor. The OT/IT network is segmented via VLANs, and sensor data flows through a secure gateway into AWS. The database workload is highly variable: during a production shift, thousands of writes per second occur, but during weekends and planned maintenance windows the database is completely idle for 60+ hours. The operations director has escalated to the architecture team that the current always-on Amazon RDS for MySQL instance is consuming 40% of the monthly AWS budget even during idle periods. The board has mandated a cost governance review and requires the architecture team to present a solution that eliminates charges during idle periods while maintaining full MySQL compatibility for existing application queries. As the solutions architect presenting to the Vanthorpe board, which database design best meets the requirement to eliminate costs during idle periods while preserving MySQL compatibility?

    • A. Migrate the workload to Amazon DynamoDB on-demand capacity mode, which scales to zero during idle periods and charges only for consumed read/write capacity units.
    • B. Replace the RDS for MySQL instance with Amazon Aurora Serverless v2 configured with a minimum Aurora Capacity Unit (ACU) setting of 0, enabling the cluster to pause and incur no compute charges during the 60+ hour idle windows.
    • C. Enable Multi-AZ on the existing RDS for MySQL instance and use Reserved Instance pricing to reduce the per-hour cost, then schedule automated snapshots to replace the standby during idle periods.
    • D. Deploy an Amazon RDS for MySQL instance with a db.t4g.micro instance class and enable automatic minor version upgrades to reduce licensing overhead during idle periods.
    View answer & explanation

    Correct answer: B. Replace the RDS for MySQL instance with Amazon Aurora Serverless v2 configured with a minimum Aurora Capacity Unit (ACU) setting of 0, enabling the cluster to pause and incur no compute charges during the 60+ hour idle windows.

    [Domain 4 - Design Cost-Optimized Architectures, Subtopic 4 - Design Cost-Optimized Architectures - Area 1 - Serverless DB] Core Concept: Design cost-optimized database solutions. requires applying serverless db in the context of the scenario, consistent with the official (ISC)² exam outline. Correct Answer (B): Replace the RDS for MySQL instance with Amazon Aurora Serverless v2 configured with a minimum Aurora Capacity Unit (ACU) setting of 0, enabling the cluster to pause and incur no compute charges during the 60+ hour idle windows. Why the other options are incorrect: Option A: DynamoDB is a NoSQL key-value and document store; it does not support MySQL-compatible SQL queries, so migrating to DynamoDB would require significant application refactoring that contradicts the MySQL compatibility requirement. Option C: Enabling Multi-AZ increases costs by adding a standby replica, and Reserved Instance pricing still charges for every hour regardless of idle state — neither change eliminates costs during idle periods as the board requires. Option D: Downsizing to a db.t4g.micro reduces per-hour cost but the instance continues running and accruing charges throughout idle weekends; this does not eliminate costs during idle periods and does not address the governance mandate.

  5. Sample question 5 · Design Cost-Optimized Architectures (20%)

    Vanthorpe Industrial, a mid-sized automotive parts manufacturer in the American Midwest, operates a shop-floor analytics platform that collects sensor readings from 400 CNC machines across two facilities. The platform ingests approximately 50,000 write events per minute during a 10-hour production shift, then drops to near-zero traffic overnight and on weekends. The current architecture uses an Amazon RDS for MySQL instance (db.r5.4xlarge) running 24/7, consuming roughly $4,200 per month. The operations team has been tasked with producing a monthly cost-efficiency report for the plant director and must demonstrate measurable database spend reduction without sacrificing shift-hour throughput. The database schema is simple: time-series sensor readings with a composite primary key of machine ID and timestamp. No complex joins or stored procedures are used. As the solutions architect, which database design best meets Vanthorpe Industrial's requirement to reduce costs while maintaining shift-hour write throughput and supporting the monthly cost-efficiency reporting objective?

    • A. Migrate the sensor data to Amazon DynamoDB with on-demand capacity mode, using machine ID as the partition key and timestamp as the sort key, and eliminate the always-on RDS instance entirely.
    • B. Replace the RDS instance with Amazon Aurora Serverless v2 configured with a minimum ACU of 8 and maximum ACU of 64 to handle peak shift traffic, keeping the MySQL-compatible engine for zero schema changes.
    • C. Retain the existing RDS for MySQL instance but purchase a 3-year Reserved Instance to reduce the hourly rate, and schedule automated snapshots every 6 hours to reduce backup storage costs.
    • D. Deploy Amazon ElastiCache for Redis in front of the existing RDS instance to cache the most recent sensor readings, reducing read pressure and allowing a downsize to a db.r5.xlarge instance.
    View answer & explanation

    Correct answer: A. Migrate the sensor data to Amazon DynamoDB with on-demand capacity mode, using machine ID as the partition key and timestamp as the sort key, and eliminate the always-on RDS instance entirely.

    [Domain 4 - Design Cost-Optimized Architectures, Subtopic 4 - Design Cost-Optimized Architectures - Area 1 - Serverless DB] Core Concept: Design cost-optimized database solutions. requires applying serverless db in the context of the scenario, consistent with the official (ISC)² exam outline. Correct Answer (A): Migrate the sensor data to Amazon DynamoDB with on-demand capacity mode, using machine ID as the partition key and timestamp as the sort key, and eliminate the always-on RDS instance entirely. Why the other options are incorrect: Option B: Aurora Serverless v2 with a minimum ACU of 8 still incurs a continuous baseline cost even at zero traffic, because ACUs scale down to the configured minimum rather than to zero; this reduces but does not eliminate idle spend, and the cost reduction is less dramatic than DynamoDB on-demand for a schema with no relational requirements. Option C: A 3-year Reserved Instance commits Vanthorpe to a fixed hourly charge regardless of utilization, locking in the same idle-hour waste for three years; this approach optimizes unit price but not architectural fit, and the cost-efficiency report would still reflect significant off-shift spend. Option D: Adding ElastiCache addresses read latency and read scaling, but the scenario describes a write-heavy ingest workload (50,000 writes per minute) with no mention of read bottlenecks; caching the most recent sensor readings does not reduce write costs, and the RDS instance must still be sized for peak write throughput, yielding minimal savings.

AWS Certified Solutions Architect – Associate exam prep FAQ

What is the passing score for the AWS Certified Solutions Architect – Associate exam?
Passing scores for AWS Certified Solutions Architect – Associate are set by AWS and can change between exam versions. Check the latest official exam outline on the AWS website for the current passing standard.
How many questions are on the AWS Certified Solutions Architect – Associate certification exam?
Official AWS Certified Solutions Architect – Associate exam length and item types are defined by AWS. Review the current exam blueprint for the latest question counts and timing.
How do domain quizzes help with AWS Certified Solutions Architect – Associate exam prep?
Domain quizzes on IT Cert Ready let you practice AWS Certified Solutions Architect – Associate one blueprint domain at a time across 4 official exam domains. You get immediate feedback, detailed explanations, and can revisit weak areas before taking a full timed mock exam.
How many AWS Certified Solutions Architect – Associate practice questions are on IT Cert Ready?
IT Cert Ready includes 1,188 published AWS Certified Solutions Architect – Associate practice questions with domain quizzes, full-length mock exams, and answer explanations mapped to the official outline.