Iterator resource intensive?

Question

Hi guys,
My code is below:
HashMap documentMap = new LinkedHashMap();
          for (Iterator it = folderInfoMap.keySet().iterator(); it.hasNext();) {
               String key = (String)it.next();
               FolderInfo folderInfo = (FolderInfo)folderInfoMap.get(key);and I has been using iterator to loop thru all the Collection object many years. However, some joker in my office telling me that I shouldn't use iterator as it's resource intensive. he asked me to use for loop like below instead:
for(int x=0;x<documentMap.size();x++{
documentMap.get(x);
}and he said that the for loop is much lesser in using resources. Any ideas?
or we should convert collection to array then loop it thru like this link mention:
http://www.exampledepot.com/egs/java.util/coll_GetArrayFromVector.html
please help, Thanks !
Thanks & Regards,
Mark

Answer 1

It's not a traversal.
It's a way of describing algorithmic complexity, i.e. the running time of algorithms, ignoring constant factors, i.e. concentrating only on the 'order' of the variables. The big O stands for 'order'. For example, bubble sort is an O(N**2) operation and is therefore avoided: decent sorting algorithms are O(N*log(N)) which gives them a better growth factor as N increases.
In your case, direct access to a HashMap is independent of the number of entries so it is described as an O(1) operation ('an operation of order 1'). Direct access to an element of a linked list takes time proportional to the length of the list, N, so it is an O(N) operation. Traversing a LinkedHashMap requires N operations of O(N) each so it is O(N**2) or 'order N squared'.
The time to do an O(1) operation is constant.
The time to do an O(N) operation is a multiple of N.
The time to do an O(N**2) operation is a multiple of N squared.
So you can see how your Iterator code stacks up against the office joker's for N=1, 2, 4, 8, ... 32768, 65536, ... Every time you double N, yours doubles in time, his quadruples, because of N versus N squared.