Content Distribution & Caching Flashcards

Which CDN feature allows different versions of the same URL to be cached based on the user's geographic location?

Answer: Vary-based caching with geographic keys

Adding geographic identifiers to the cache key enables distinct cached responses per region for the same URL.

What is the primary risk of setting excessively long TTLs (e.g., 1 year) for HTML pages on a CDN?

Answer: Stale content being served after site updates

Long TTLs mean updated HTML may not reach users until caches expire, requiring manual purges for timely updates.

A CDN uses 'request coalescing' (also called request collapsing). What problem does this solve?

Answer: Prevents multiple simultaneous cache misses from all hitting origin for the same resource

Request coalescing queues duplicate in-flight requests at the edge so only one request goes to origin, preventing thundering herd.

Which metric specifically measures the percentage of requests served from CDN cache without hitting origin?

Answer: Cache hit ratio (CHR)

Cache hit ratio is the fraction of total requests served from cache, directly measuring CDN caching effectiveness.

Edge Side Includes (ESI) are primarily used to:

Answer: Assemble web pages from individually cacheable fragments at the edge

ESI allows edge servers to stitch together page fragments with different cache lifetimes, enabling partial page caching.

A CDN operator wants to ensure API responses with sensitive user data are never cached. Which Cache-Control combination is most appropriate?

Answer: no-cache, no-store, private

'no-cache, no-store, private' prevents both storage and serving of sensitive responses at any cache layer.

What is 'origin shield' (also called 'shield PoP' or 'mid-tier caching') in a CDN architecture?

Answer: A designated intermediary CDN node that aggregates cache misses before they reach origin

Origin shield is a single CDN node that acts as a proxy for all edge nodes, dramatically reducing origin traffic on cache misses.