Free SAA-C03 practice scenarios: S3 storage classes

Paying the right price for each access pattern — infrequent, unknown, archival — and automating it with lifecycle rules. Six scenarios, SAA-C03, 2026 edition.

Try 6 free S3 storage-class scenarios in the exam's own style — every answer explained, no sign-up.

Straight to the 6 free questions ↓

S3 storage classes are the cost domain’s most concrete pattern: a scenario names an access frequency, a retrieval-time tolerance and a durability need, and asks for the cheapest class that still meets all three. The classes line up along one axis — the colder the data, the cheaper the storage and the slower the retrieval.

The ladder

S3 Standard — frequently accessed, instant.
Standard-IA — infrequently accessed, still instant retrieval, cheaper storage.
One Zone-IA — same as Standard-IA but a single AZ, cheaper again — for re-creatable data where lower durability is acceptable.
Intelligent-Tiering — unknown or changing patterns; moves objects between tiers automatically with no retrieval fees.
Glacier Flexible Retrieval / Deep Archive — archives, retrieval in minutes to hours; Deep Archive is the cheapest storage of all, with retrieval up to ~12 hours.

Read the requirement’s retrieval tolerance first: “immediately” rules out Glacier; “12 hours is fine, cheapest” points straight at Deep Archive.

Intelligent-Tiering vs a lifecycle policy

Two ways to handle changing data, and the exam distinguishes them. Intelligent-Tiering is for when the pattern is unknown — it decides per object automatically. A lifecycle policy is for when the pattern is known and time-based — “move to IA after 30 days, Glacier after 90” — transitioning objects by age with no manual work. Unknown pattern → Intelligent-Tiering; known schedule → lifecycle policy.

The trap: IA is not always cheaper

The most-tested subtlety is that Standard-IA is not automatically the cheaper choice. It carries a per-GB retrieval fee and a 30-day minimum storage duration, so frequently read or short-lived objects rack up retrieval charges and pay for storage they never use — often more than Standard. IA only wins for data that is genuinely infrequent and kept a while. When a stem describes small, hot, short-lived objects, the cheaper storage rate is a false economy.

How to use the six scenarios below

Answer each one before opening the explanation, and when you miss, name the three requirements that decided it — access frequency, retrieval tolerance, durability need — plus any hidden retrieval fee or minimum duration. Every explanation states the deciding constraint and dismantles the distractors, which are usually a class that is cheaper to store but wrong on retrieval time or on total cost. For the domain overview, see Design Cost-Optimized Architectures.

The S3 storage classes ordered by access frequency and cost. S3 Standard for frequently accessed data, Standard-Infrequent Access for rarely accessed data needing immediate retrieval, One Zone-Infrequent Access for the same but re-creatable data in a single Availability Zone at lower cost, Intelligent-Tiering for unknown or changing access patterns with automatic tiering and no retrieval fees, and the Glacier classes for archives, with Glacier Deep Archive the cheapest at retrieval times up to twelve hours. A lifecycle policy transitions objects from Standard to Infrequent Access and on to Glacier over time.

6 free SAA-C03 practice questions

Answers and explanations — no email wall

SAA-C03 Question 1 of 6

A company keeps documents that are accessed only a few times a year, but when they are needed they must be retrieved immediately. It wants to pay less than S3 Standard. Which storage class fits?

Answer: A — S3 Standard-Infrequent Access (Standard-IA).

Standard-IA costs less per gigabyte than Standard and still offers immediate, millisecond retrieval — the right fit for rarely accessed data that must be available instantly. Why not the others: S3 Standard is more expensive and aimed at frequent access. The Glacier classes are cheaper for storage but impose retrieval delays of minutes to hours, which breaks the 'immediately' requirement. Infrequent access with instant retrieval is Standard-IA.

SAA-C03 Question 2 of 6

An application stores re-creatable processing logs that are accessed infrequently. Losing them would be acceptable because they can be regenerated, and the company wants the lowest storage cost for infrequent access. Which class is most appropriate?

Answer: A — S3 One Zone-Infrequent Access (One Zone-IA).

One Zone-IA stores infrequently accessed data in a single Availability Zone, costing less than Standard-IA — ideal when the data is re-creatable, so the lower durability of one AZ is an acceptable trade. Why not the others: Standard-IA stores across multiple AZs and costs more, which is unnecessary for data you can regenerate. Standard is dearer still. Glacier Deep Archive is for archives with long retrieval times, not infrequently but readily accessed logs. Re-creatable, infrequent, cheapest is One Zone-IA.

SAA-C03 Question 3 of 6

A data set has unknown and changing access patterns — some objects become hot, others go cold unpredictably — and the team does not want to manage transitions or risk retrieval fees. Which storage class fits best?

Answer: A — S3 Intelligent-Tiering.

Intelligent-Tiering monitors each object's access and moves it between frequent and infrequent tiers automatically, with no retrieval fees and no operational effort — purpose-built for unknown or changing patterns. Why not the others: putting unpredictable data in Standard-IA risks retrieval charges every time a 'cold' object turns hot. Standard never optimises and overpays for the cold objects. A blanket lifecycle rule to Glacier would archive objects that may still be accessed, adding retrieval delays. Unknown, changing access is Intelligent-Tiering.

SAA-C03 Question 4 of 6

A company must retain compliance records for seven years. The records are almost never accessed, and when they are, a retrieval time of up to 12 hours is acceptable. The priority is the lowest possible storage cost. Which class fits?

Answer: A — S3 Glacier Deep Archive.

Glacier Deep Archive is the lowest-cost S3 storage, designed for long-term archives accessed rarely, with retrieval times up to around 12 hours — matching 'seven years, almost never accessed, 12 hours is fine, cheapest'. Why not the others: Glacier Flexible Retrieval is cheap but costs more than Deep Archive and exists for faster retrievals the requirement does not need. Standard-IA and Intelligent-Tiering keep data instantly available at much higher storage cost, which the 'lowest possible cost, 12 hours acceptable' requirement makes wasteful. Cheapest long-term archive is Deep Archive.

SAA-C03 Question 5 of 6

A company wants objects to move automatically from S3 Standard to Standard-IA after 30 days and then to S3 Glacier after 90 days, without manual intervention. How should a solutions architect implement this?

Answer: A — An S3 Lifecycle policy with transition rules based on object age.

An S3 Lifecycle policy transitions objects between storage classes automatically based on their age — Standard to Standard-IA at 30 days, then to Glacier at 90 — with no manual work. Why not the others: a custom Lambda copier reinvents a native feature and adds cost and failure modes. Changing classes by hand each month is exactly the manual toil the requirement forbids. Versioning preserves object versions and has nothing to do with tiering by age. Age-based transitions are an S3 Lifecycle policy.

SAA-C03 Question 6 of 6

A team plans to store many small objects that are read frequently and kept for only a few days, and proposes Standard-IA to save money. Why is this likely to cost MORE than S3 Standard?

Answer: A — Standard-IA has a per-GB retrieval charge and a 30-day minimum storage duration, which penalise frequent access and short lifetimes.

Standard-IA is cheaper to store but adds a per-GB retrieval fee and bills a 30-day minimum even for objects deleted sooner — so frequently read, short-lived data incurs retrieval charges and pays for storage it never uses, often exceeding Standard's cost. Why not the others: Standard-IA has the same eleven-nines durability across multiple AZs, so no double-storing is needed. It can store small objects (though a per-object minimum size applies for billing). Its per-GB storage rate is lower, not higher — the extra cost comes from the retrieval fee and minimum duration, not the storage rate. IA is only cheaper for genuinely infrequent, longer-lived data.

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 6 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 does S3 storage-class selection show up on the SAA-C03?

In the cost-optimization domain, usually as a scenario giving an access frequency, a retrieval-time tolerance and a durability need, and asking for the cheapest class that still meets them. The trap is choosing a cheaper storage rate that carries retrieval fees or minimum durations the workload will actually hit.

When is Intelligent-Tiering the right answer?

When the access pattern is unknown or changing and you want automatic optimisation without managing transitions or paying retrieval fees for objects that turn hot again. If the pattern is known, a specific class plus a lifecycle policy is usually cheaper.

Are these real exam questions?

No. They are original scenarios written for this page in the exam's style and difficulty. Reproducing real exam items violates the AWS certification agreement and can cost a candidate their certification — ours come with the reasoning attached instead.

What is the one-line takeaway?

Frequent → Standard; infrequent but instant → Standard-IA (One Zone-IA if re-creatable); unknown pattern → Intelligent-Tiering; archive → Glacier, Deep Archive for the cheapest; automate transitions with a lifecycle policy.

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.