AWS Solutions Architect Associate SAA-C03 Sign in Try 10 free questions

Free SAA-C03 practice questions: Cost-Optimized Architectures

The smallest domain by weight, but the one whose vocabulary — Spot, Savings Plans, storage classes — shows up in every other domain's answer options — SAA-C03, 2026 edition.

Here are 7 cost-optimization questions in the exam's own scenario style — free, with every answer explained, no sign-up.

Straight to the 7 free questions ↓

Design Cost-Optimized Architectures is the smallest of the four SAA-C03 domains at 20% of the scored content, and the most commonly misread. Candidates expect arithmetic — which option costs less — and the exam almost never asks that. It asks which option costs less for this workload, and the distance between those two questions is the entire domain.

Cost optimization is constraint-matching, not penny-pinching

Here is the pattern that makes SAA-C03 cost questions feel slippery: several of the four options genuinely save money somewhere. Spot is cheaper than On-Demand. Glacier is cheaper than Standard. A Savings Plan beats list price. The wrong answers are rarely expensive options — they are discounts whose conditions this particular workload cannot meet.

Every discount AWS offers is payment for something you give up: flexibility, immediacy, or certainty. So every stem contains one deciding constraint — can the workload survive an interruption, how quickly must the data come back, will usage hold steady long enough to commit. Find that constraint first, before you look at the options, and most questions in this domain collapse to a single surviving answer.

The compute purchasing ladder

The diagram above shows the ladder the exam keeps climbing. On-Demand is the baseline: no commitment, full list price, and the correct answer only when the stem describes spiky, unpredictable, or short-lived usage that nothing else fits. Savings Plans trade a one- or three-year hourly-spend commitment for discounts of up to roughly 66% (Compute Savings Plans) while staying flexible across instance families, sizes, and even compute services. Reserved Instances commit harder — to instance attributes, not just spend — and reach up to roughly 72%. Spot sits at the deep end, up to roughly 90% off, because you agree that AWS can reclaim the capacity with two minutes’ notice.

The question that unlocks Spot is always the same: can this workload die mid-run and carry on? Checkpointing, stateless workers, fault-tolerant batch processing, containerised jobs that reschedule — these phrases are the exam telling you interruption is survivable. A steady 24/7 baseline points to a commitment instead, and the Savings Plan versus Reserved Instance choice then turns on how much flexibility the future needs: if the stem hints that instance types or Regions might change, the flexible commitment wins.

The S3 storage ladder and the retrieval question

Storage classes run the same trade in a different currency. From S3 Standard down through Standard-IA, Glacier Instant Retrieval, Glacier Flexible Retrieval, and Glacier Deep Archive, the per-GB storage price falls — and two other numbers rise: the cost of retrieving an object and the time a retrieval takes.

The exam’s decider is nearly always the retrieval-time requirement, stated plainly in the stem. Milliseconds keeps you on the instant side of the ladder — the only question left is how infrequent access is, which separates Standard, Standard-IA, and Glacier Instant Retrieval. Minutes to hours opens Glacier Flexible Retrieval. Twelve hours of patience unlocks Deep Archive, the cheapest storage AWS sells.

Two automation tools sit on top of the ladder, and the exam distinguishes them sharply. Lifecycle policies are for access patterns you know: data that is hot for 30 days, cool for 60, and archival for seven years descends the ladder on a schedule. S3 Intelligent-Tiering is for patterns you do not know: it watches each object, moves it between tiers as usage changes, and charges no retrieval fees. “Access patterns are unknown” in a stem is as close to an answer key as this exam gets.

The tooling trio: analyse, alert, raw detail

Three cost-management services show up as each other’s distractors, and each answers a different verb. Cost Explorer analyses: interactive charts, filters, forecasts — the tool for understanding spend that has happened. AWS Budgets alerts: you define a cost or usage threshold and it notifies you — or triggers actions — when actual or forecast spend crosses it. The Cost and Usage Report is the raw detail: hourly, resource-level line items with tag columns, delivered to S3 for Athena or a data warehouse to consume. Match the stem’s verb — understand, warn, export — and the trio stops being confusable.

A fourth tool answers the verb recommend: AWS Compute Optimizer applies machine learning to CloudWatch utilisation history and names the instance type a workload should be running on. When a stem asks for specific right-sizing recommendations rather than a flag on waste, Compute Optimizer is the answer the exam wants.

The hidden cost topic: data transfer

The quietest recurring cost question involves no purchasing options at all. Instances in private subnets reaching S3 through a NAT gateway pay a per-GB data-processing charge on every byte — a cost many architectures carry without noticing. The exam’s recurring right answer is the gateway VPC endpoint: for S3 and DynamoDB, it routes the traffic over the AWS network with no hourly fee and no per-GB charge, and the NAT gateway’s processing fee on that traffic disappears entirely. It is worth internalising as a pattern, not a fact — one of the few exam answers that removes a cost rather than discounting it, which is why the question keeps coming back.

How to use the questions below

Read each stem and name the deciding constraint — interruption tolerance, retrieval time, commitment horizon, exposure requirement — before you look at the options. Then answer, and read the explanation past the correct letter: each wrong option is wrong for a specific, nameable reason, and several would save money in a different scenario. Learning which scenario each one belongs to is the skill this domain scores — and the one that quietly earns marks across the rest of the exam too.

Four columns comparing the EC2 purchasing options as a ladder of deepening discounts. On-Demand carries no commitment at full list price and serves as the baseline. Savings Plans commit to an hourly spend for one or three years, remain flexible across compute, and reach up to roughly 66% off. Reserved Instances commit to specific instance attributes for one or three years and reach up to roughly 72% off. Spot Instances are interruptible with two minutes' notice and reach up to roughly 90% off. A bar beneath each option grows with its discount depth, and a note beneath the columns states that the discount never decides the answer — the workload's tolerance for commitment or interruption does.

7 free SAA-C03 practice questions

Answers and explanations — no email wall
SAA-C03 Question 1 of 7

A data analytics company runs a nightly batch job on Amazon EC2 that takes several hours, writes a checkpoint to Amazon S3 every few minutes, and can resume from the last checkpoint after any failure. The job must finish by morning, but which specific instances run it does not matter. Which purchasing option is the MOST cost-effective?

Answer: B — Spot Instances

The stem hands you the unlock in plain sight: checkpoints plus resume-after-failure means the workload tolerates interruption, and interruption tolerance is exactly what Spot pays for — discounts of up to around 90% in exchange for AWS reclaiming the capacity with two minutes' notice. Why not the others: On-Demand pays full list price for a flexibility this job does not need. A Reserved Instance and a Savings Plan both reward a steady, committed level of usage — a nightly batch job has no 24/7 baseline to commit to, and even where it did, both discounts are shallower than Spot's. When a workload can die mid-run and carry on, Spot wins; the other options exist for workloads that cannot.

SAA-C03 Question 2 of 7

A media company stores finished video projects in Amazon S3. After delivery, a project is accessed about once per quarter, but when an editor requests a file it must be available within milliseconds. Storage cost should be as low as possible. Which storage class meets these requirements?

Answer: C — S3 Glacier Instant Retrieval

Two constraints decide this: access roughly quarterly (very infrequent) and retrieval in milliseconds. S3 Glacier Instant Retrieval is built for exactly that pair — the lowest storage price of any class that still returns objects with millisecond latency, designed for data touched about once a quarter. Why not the others: S3 Standard delivers milliseconds but at the highest storage price, paying for a frequent-access profile this data does not have. Standard-IA also delivers milliseconds, but it is priced for monthly-ish access and stores at a higher rate than Glacier Instant Retrieval, so it is beaten on cost. Glacier Deep Archive is far cheaper at rest but takes hours to restore — the millisecond requirement eliminates it before price is even discussed.

SAA-C03 Question 3 of 7

Application log files in Amazon S3 are analysed heavily for 30 days after creation, referenced only occasionally for the next 60 days, and must then be retained for seven years for compliance, during which access is not expected. What is the MOST cost-effective storage strategy?

Answer: A — Store logs in S3 Standard and add a lifecycle policy that transitions them to S3 Standard-IA after 30 days and to S3 Glacier Deep Archive after 90 days.

The access pattern is known and it ages in stages — hot, cooling, cold — which is precisely the case S3 lifecycle policies exist for: each stage gets the class its access frequency deserves, automatically. Why not the others: Standard-IA from day one puts a per-GB retrieval charge and a 30-day minimum-storage class under the heaviest analysis phase — infrequent-access pricing punishes frequent access. Deep Archive from day one makes the first 30 days of analysis effectively impossible, since restores take hours. Versioning is a durability and recovery feature, not a cost lever — and permanently deleting anything after 90 days violates the seven-year retention requirement outright.

SAA-C03 Question 4 of 7

A company is migrating a large file share with millions of objects to Amazon S3. Access patterns are unknown and vary from object to object, and the company wants to minimise storage cost without any operational overhead or retrieval fees. Which storage class should it choose?

Answer: A — S3 Intelligent-Tiering

The phrase 'access patterns are unknown' is the exam's standing invitation to answer S3 Intelligent-Tiering: it monitors each object individually, moves it between access tiers as its usage changes, and charges no retrieval fees — the small monitoring charge is the price of not having to know the pattern. Why not the others: Standard-IA assumes you already know the data is infrequently accessed; objects that turn out to be hot pay retrieval fees on every read. One Zone-IA adds the same problem plus a genuine availability trade-off, holding data in a single Availability Zone. The fixed lifecycle policy hard-codes a guess — any object still in active use after 30 days would land in Glacier Flexible Retrieval, behind minutes-to-hours restores and retrieval charges. Automation beats assumption when the pattern is unknown.

SAA-C03 Question 5 of 7

A FinOps team needs to load hourly, resource-level AWS billing data — including cost allocation tags — into its own data warehouse and query it with Amazon Athena for departmental chargeback. Which source should the team use?

Answer: C — AWS Cost and Usage Report

Hourly granularity, resource-level line items, tag columns, delivery to Amazon S3, Athena on top — every clause of the stem describes the Cost and Usage Report, the most detailed billing record AWS produces and the raw material for any serious chargeback pipeline. Why not the others: Cost Explorer is the analysis tool — interactive charts, filters and forecasts — but it is a console experience with limited granularity, not a raw data feed for a warehouse. AWS Budgets answers a different verb entirely: it alerts when actual or forecast spend crosses a threshold you define. The Pricing Calculator looks forwards, estimating workloads that do not exist yet — chargeback is about money already spent. Analyse, alert, estimate — none of them is 'export the raw detail'.

SAA-C03 Question 6 of 7

A company suspects that many of its several hundred EC2 instances are over-provisioned. It wants recommendations for specific smaller or different instance types, derived from machine-learning analysis of historical CloudWatch utilisation metrics. Which service provides this?

Answer: A — AWS Compute Optimizer

Right-sizing with named target instance types, driven by machine-learning analysis of utilisation history, is the definition of AWS Compute Optimizer — it looks at how an instance actually behaves and recommends the type it should be, including across instance families. Why not the others: Trusted Advisor's cost-optimisation checks will flag a low-utilisation instance as waste, but flagging is not the same as recommending a specific replacement type from metric analysis — when the stem says 'machine learning' and 'specific instance types', the exam means Compute Optimizer. CloudWatch supplies the raw utilisation metrics but draws no conclusions from them; you would be doing the analysis yourself. AWS Config records configuration state and compliance over time — it knows what an instance is, not whether it is the right size.

SAA-C03 Question 7 of 7

EC2 instances in private subnets transfer several terabytes per day to Amazon S3 buckets in the same AWS Region, routed through a NAT gateway. Data-processing charges have become significant. How can the company reduce this cost WITHOUT exposing the instances to the internet?

Answer: A — Create a gateway VPC endpoint for S3 and route the S3 traffic through it.

A gateway VPC endpoint for S3 carries the traffic over the AWS network with no hourly charge and no per-GB processing charge — the NAT gateway's per-GB fee on those terabytes simply disappears, and the instances stay private. This is the exam's recurring data-transfer answer, and it recurs because it is one of the few options that removes a cost rather than discounting it. Why not the others: a second NAT gateway doubles the hourly cost and splits the same per-GB charges across two devices — the total does not shrink. Public subnets violate the stated security constraint before cost is even considered. S3 Transfer Acceleration speeds up long-distance transfers over edge locations and adds a per-GB charge of its own — it makes this bill larger, not smaller.

That is exactly how every question in the course works — answer, explanation, why-not. The real set continues in the practice player: 10 free questions, no sign-up.

Continue with the 10 free questions →

Those 7 questions were the start.

The exam does not test whether you recognise a term — it tests whether you can rule out three plausible answers under time pressure. That is what the explanations above are for, and there are 780 more questions built exactly like them.

Collecting questions yourself

  • Scattered across forums, of unknown age
  • Answer keys without reasoning
  • No idea which domain you are weak in

Practising with a system

  • 780 questions in 12 full tests, SAA-C03 (2026)
  • Every option explained — including the wrong ones
  • Readiness per exam domain, and drills for your weakest
Start free — 10 questions

Straight into the player. No account, no email.

Frequently asked questions

How much of the SAA-C03 exam is Design Cost-Optimized Architectures?
It is domain 4 and accounts for 20% of the scored content per the official exam guide — the smallest of the four domains, but its concepts (purchasing options, storage classes, data-transfer costs) also appear inside questions scored under other domains. Our simulator draws to these official domain weights, so a full-length run gives you a realistic share of cost questions.
Why are cost questions really trade-off questions?
Because on the SAA-C03 exam, several options usually save money somewhere — Spot is cheaper than On-Demand, Glacier is cheaper than Standard, a commitment beats list price. The question is never which discount is deepest but which discount's conditions the workload can meet: can it survive interruption, how long can retrieval take, will usage stay steady enough to commit. The deciding constraint is always stated in the stem; the skill is spotting it.
Are these real exam questions?
No. They are original questions written in the style and difficulty of the exam. Sharing real exam items violates the AWS certification agreement and can cost a candidate their certification — which is why nobody serious does it. What actually prepares you is the reasoning attached to every answer here, which real items never come with anyway.

Updated for SAA-C03 (July 2026). The sample questions above are our own work in the style of the exam — not real exam items. The exam itself is set and marked by the certification body.