We were essentially doing the lose-lose algorithm. It's a functioning hash table but there's a lot of room for improvement.