[dredmorbius]

Requests follow an inverse log relationship of request frequency.

A little caching buys you tremendous savings in regeneration or back-end fetching.