Incremental Static Regeneration

Incremental static regeneration is a hybrid of static and dynamic rendering: a page is generated once and served from cache, and after a configured interval the next request triggers a background regeneration while still being served the stale copy. Visitors always get a fast response; the content converges on fresh.

It is the standard answer for pages built from data that changes on a scale of minutes to hours — a feed of repository activity, a list of recent posts, anything drawn from a third-party API. It removes per-request latency and, just as importantly, removes the API from the critical path: if the upstream is down, the cached page still serves.

Two things are worth designing deliberately. The interval encodes how stale you are willing to be, and different data on one page may deserve different intervals. Invalidation is the escape hatch for content that must update immediately — tagging cached entries and purging by tag beats waiting for a timer. See Cache Invalidation.

The failure mode to watch for is caching that is not request-aware. A value memoised in module scope looks like a cache and behaves like one within a single process, but it does not participate in the platform's revalidation and does not invalidate across instances — which is a correctness problem, not a performance one.