I need some help with the following problem from Introduction to Algorithms by Cormen, Leiserson, Rivest, Stein:
Consider an ordinary binary min-heap data structure with $n$ elements that supports the instructions INSERT and EXTRACT-MIN in $O(\lg n)$ worst-case time. Give a potential function $\Phi$ such that the amortized cost of INSERT is $O(\lg n)$ and he amortized cost of EXTRACT-MIN is $O(1)$, and show that it works.
I'm not sure how to define the potential here. Especially since everything looks so "symmetric".
I mean, I get that one can only do as many EXTRACT-MIN operations as one does INSERT operations, and so if we have $k$ INSERT operations and $\ell$ EXTRACT-MIN operations total, then the time this takes is surely not bigger than doing the INSERT first and the EXTRACT-MIN operations later. And so I guess, the time for all operations is
$$ \begin{align} t(k+\ell) &\le (\lg 1 + \lg 2 + \dots + \lg k) + (\lg k + \lg (k-1) + \dots + \lg (k-\ell+1)) \\ &= 2\lg k! - \lg \ell! \\ &= O(k\lg k) \end{align} $$
Hence we can assign $O(\lg k)$ to every insert operation and $O(1)$ to every EXTRACT-MIN operation (?).
So the result at least makes some sense. But I still don't know how to do it formally using a potential.
Thanks for your help! =)
