What describes the cache invalidation on writes strategy?

What describes the cache invalidation on writes strategy?

• A. Delete the cache entry immediately after the database record is updated.
• B. Refresh the cache only after a read misses.
• C. Update the cache and database atomically.
• D. Keep the cache stale to maximize performance.

Answer: (A)

Cache invalidation on writes focuses on ensuring that cached data isn’t allowed to become stale after an update. The most straightforward and reliable way is to delete the cache entry as soon as the corresponding database record is updated. Once the cache is purged, the next read fetches the fresh value from the database and then repopulates the cache. This keeps reads correct with minimal complexity and avoids serving outdated data, even though the next read will incur a cache miss and require a fresh fetch. Why the other ideas don’t fit as well: refreshing the cache only after a read misses can still expose clients to stale data in the window after a write and before a read occurs. Updating the cache and database atomically is a different pattern that hinges on cross-system transactional guarantees and isn’t the same as simply invalidating on writes. Keeping the cache intentionally stale prioritizes performance over correctness, which defeats the goal of ensuring up-to-date information.

Cache invalidation on writes focuses on ensuring that cached data isn’t allowed to become stale after an update. The most straightforward and reliable way is to delete the cache entry as soon as the corresponding database record is updated. Once the cache is purged, the next read fetches the fresh value from the database and then repopulates the cache. This keeps reads correct with minimal complexity and avoids serving outdated data, even though the next read will incur a cache miss and require a fresh fetch.

Why the other ideas don’t fit as well: refreshing the cache only after a read misses can still expose clients to stale data in the window after a write and before a read occurs. Updating the cache and database atomically is a different pattern that hinges on cross-system transactional guarantees and isn’t the same as simply invalidating on writes. Keeping the cache intentionally stale prioritizes performance over correctness, which defeats the goal of ensuring up-to-date information.