Which is a core step in the five-step caching strategy?

Which is a core step in the five-step caching strategy?

• A. Cache everything.
• B. Ignore invalidation concerns.
• C. Choose a write-through approach for all data.
• D. Set an eviction policy (LRU plus TTL).

Answer: (D)

In caching, managing what stays in the cache and for how long is essential to balance speed with correctness. Setting an eviction policy, such as LRU combined with TTL, is the step that directly governs how the cache decides which items to remove when space is needed and how long entries remain valid. LRU keeps frequently accessed data in memory by evicting items that haven’t been used recently, while TTL automatically expires entries after a set time to prevent serving very stale data. Together, they control cache size, hit rate, and data freshness, making this step fundamental to a practical caching strategy. Choosing to cache everything isn’t realistic or safe for memory and freshness reasons. Ignoring invalidation concerns risks serving stale data. Adopting a write-through approach for all data is a policy choice about writes, not about how the cache itself manages capacity and expiration. The eviction policy with LRU and TTL is the mechanism that actually enables effective and reliable caching under real workloads.

In caching, managing what stays in the cache and for how long is essential to balance speed with correctness. Setting an eviction policy, such as LRU combined with TTL, is the step that directly governs how the cache decides which items to remove when space is needed and how long entries remain valid. LRU keeps frequently accessed data in memory by evicting items that haven’t been used recently, while TTL automatically expires entries after a set time to prevent serving very stale data. Together, they control cache size, hit rate, and data freshness, making this step fundamental to a practical caching strategy.

Choosing to cache everything isn’t realistic or safe for memory and freshness reasons. Ignoring invalidation concerns risks serving stale data. Adopting a write-through approach for all data is a policy choice about writes, not about how the cache itself manages capacity and expiration. The eviction policy with LRU and TTL is the mechanism that actually enables effective and reliable caching under real workloads.