SAA-C03 Practice Question 9: S3 Intelligent Tiering Unknown Access

S3 Intelligent-Tiering exists for precisely this case: data with unknown, changing, or unpredictable access patterns. For a small per-object monitoring and automation fee it moves each object individually between the Frequent Access tier, the Infrequent Access tier (after 30 consecutive days without access), and the Archive Instant Access tier (after 90), and it moves an object straight back to Frequent Access the moment it is read. All three of those tiers are low-latency, so the millisecond requirement holds, there are no retrieval fees, and there is no minimum storage duration to be caught out by. Since every object here is over 1 MB, none fall under the 128 KB threshold below which objects are not monitored.

Option B punishes exactly the objects the team reads often. S3 Standard-IA and S3 One Zone-IA both charge per-GB retrieval fees and impose a 30-day minimum storage duration, so a dataset that is still being read weekly becomes *more* expensive after the transition, not less. S3 One Zone-IA also drops to a single Availability Zone, which is a durability posture change nobody asked for.

Option C keeps millisecond access (Glacier Instant Retrieval is the archive class that does), but it carries a 90-day minimum storage duration and per-GB retrieval fees. Objects still in active use at day 31 would incur retrieval charges on every read and an early-deletion charge if replaced, and a flat 30-day rule cannot tell an active dataset from a dead one.

Option D is the manual version of what Intelligent-Tiering automates, and it is the highest-effort answer on offer. S3 Storage Lens reports aggregate storage metrics and trends; it is not a per-object last-accessed feed, so the team would be guessing anyway, every month, forever.

Reference: Understanding and managing Amazon S3 storage classes