CareerCup

TRIE is the best option to implement a cell phone contact lists, though they consume a lot of memory...

tries should be super fast, but they use a lot of memory.

I am assuming all the entries reside in some database.
If memory is not an issue, once a user types a letter, say, "p", we can create a hashtable with all names starting with "p". Then as the user types further filter the hashtable.
Well the worst case would be all names start with "p" :)).

i think this question has already been asked in other threas...i guess the best solution from the context of efficiency is to use suffix trees and hence TRIE...they do exactly the same thing as what we are looking for...hope this helps. thankx

Wht is TRIE ?? is it different from tree data structure??

@ above ^
Yes, it is unique kind of data structure.
You can see the wiki page for the trie thing.

Directed acyclic word graph would be more efficient in terms of memory.