3.3Standard
Responses are cached to reduce inference cost.
Prevent cross-tenant leakage.
Choose two designs.
× Incorrect
Include tenant, authorization, model, prompt, and data versions in cache keys and manage TTL, invalidation, privacy, encryption, hit rate, and staleness.
Detailed explanation
CorrectUse TTL, invalidation, privacy controls, encryption, and stale-result monitoring
Cost and leakage are controlled.
CorrectInclude tenant, permission, model, prompt, and data versions in the cache key
Results are less likely to mix.
IncorrectThe same text question has one answer for every user
Context and permissions differ.
IncorrectNever expire or delete a cached result
Old or personal data remains.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 3.3・5.1のキャッシュ、データ分離、保持を確認する。
Official references
What to confirm
コスト削減のキャッシュが権限混同や古い回答を生まないよう設計できる。
Key points
- Cache key
- TTL
- Invalidation
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Cache key
The set of attributes that determines whether a stored result may be reused.
TTL
A time-to-live after which a cached item is treated as expired.
Question IDAWS-AIF-462
EnvironmentAWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
Last technical review2026-08-20
Report an error or rights issue