Caching mindful of computation/storage costs

 Humans repeat stuff. Caching helps.
 .
 Normal caching policies like LRU aren't well suited for analytic computations
 where both the cost of recomputation and the cost of storage routinely vary by
 one million or more.
 .
 Cachey tries to hold on to values that have the following characteristics
 .
  1. Expensive to recompute (in seconds)
  2. Cheap to store (in bytes)
  3. Frequently used
  4. Recenty used