A lakehouse holds seven years of transaction history, of which only the last quarter is queried regularly. How should the data be placed?
Select an answer to reveal the explanation.
Short Explanation
You keep the actively queried recent data on a performance tier and move the older, rarely accessed history to a lower-cost tier. Hot data on fast storage, cold history on cheap storage. It’s the obvious answer and the right one—the trap is that both extremes look decisive and both are wrong.
Full Explanation
The correct answer is the choice that keeps the actively queried recent data on a performance tier and moves the older, rarely accessed history to a lower-cost tier. Tiering matches placement to access pattern. The recent quarter earns its performance tier through frequent access; six-and-three-quarter years of rarely touched history does not, and moving it captures most of the available saving with no practical impact.
Keeping all seven years on the fastest tier so any query is fast is incorrect because paying premium rates for data that is almost never read wastes money for a benefit nobody experiences.
Moving all seven years, including the current quarter, to archival storage to minimise cost is incorrect because archiving the actively queried quarter makes routine work slow or subject to retrieval delays.
Assigning tiers by file extension, so all files of the same type share a tier, is incorrect because file extension describes format, not how often the data is read—the only thing that should drive the decision.
For the exam, drive tiering from access pattern, retention need and performance requirement.