Rarely read data older than 90 days should cost less, automatically. What do you configure?
Why D is the answer
An S3 Lifecycle rule to a colder storage class
Lifecycle rules move objects between storage classes based on age, with no code.
In detail
- A Lifecycle rule moves data to a colder class automatically.
- Check retrieval costs before you choose a class.
Use case
An illustrative scenario, not a real incident.
Logs are rarely read after 90 days. A Lifecycle rule moves them to a colder storage class, with no code.
See it in different ways
stateDiagram-v2 [*] --> Standard Standard --> InfrequentAccess: 30 days InfrequentAccess --> Glacier: 90 days Glacier --> Expired: retention ends Expired --> [*]
stateDiagram-v2 [*] --> RawCsv RawCsv --> Parquet: columnar, compressed Parquet --> Partitioned: by date Partitioned --> Compacted: fewer small files Compacted --> [*]
stateDiagram-v2 state "Recent objects" as s0 state "Colder storage" as s1 state "Expired" as s2 [*] --> s0 s0 --> s1: after 90 days s1 --> s2: if a rule allows s2 --> [*]
Key point: Lifecycle rules act on object age.
Why the other options are wrong
- A. A Glue crawler
- A crawler updates a catalog.
- B. A larger Redshift cluster
- A larger cluster costs more.
- C. An EventBridge schedule that deletes data
- Deleting loses the data.