For loop vs Iterator

Question

I'm writing an application in which performance will be quite important and was looking into my collection classes. I recently saw the move towards using Iterator instead of for loops, while loops to retrieve objects from a List. I wanted to test to see which performed better and I saw that for about 1,000,000 Integer objects, an Iterator was 100ms slower in time in both Vector and ArrayList than a for loop using gets.
1) Is the main cost the creation of the Iterator object?
2) If I'm not worried about manipulation of my List while I'm retrieving objects and I will always be using a List in which I can always pull indexes (instead of say using HashSet)... is there any reason to use Iterator instead of a for loop?
-- I read the recent tech tip on this and was just wondering if anyone had any other ideas of why to choose Iterator over for/while loops.
Thanks.

Answer 1

The purpose of Iterator
The purpose of the Iterator pattern is to handle the traversal of a collection of objects.
The advantages of Iterator has over other straightforward ways
The direct access methods provided by some collection classes may not be as efficient or simple as using iterator pattern. And because the iterator keeps its own state, multiple traversals can be performed on a single collection simultaneously.
An iterator hides the internal representation of the collection from the classes that need to access the contents of these collections.
An Iterator pattern provides a consistent and sometimes efficient way to access the objects or values in some collections.
What problems can arise if collection is changing?
While using iterator, the changes of a collection such as that adding new elements or deleting old elements can result in inconsistency of the data, for example some objects may be missed or accessed more than once.
How can they be avoided?
A simple way is to make a copy of the collection and work on this copy but this can be very costly in terms of memory and time.
Design a method that notifies the iterator whenever the collection is changed, and as soon as the change happens the iterator class throws an exception or restart whichever appropriate.
A better way would be implements methods that monitor the changes and adjust the iterator and the results so they are consistent to the contents of the collection at any time.
There are no general solutions to this problem and under different circumstance the proper solutions vary, however it is a good practice to taking possible changes of the source collections into account while designing the iterator, and it may be worth to let the user aware of the occurrence of any changes.