I use two structures for persistent data: Extendible hashing for unordered data and skip lists for ordered data. Extendible hashing is great for looking up an o

I use two structures for persistent data: Extendible hashing for unordered data and skip lists for ordered data. Extendible hashing is great for looking up an object given its UID. Skip lists are easier than B-trees (to me) for everything else. I also use skip lists occasionally for in-memory data because they're just so darn handy. I wish both were better known.