Global Counter in hadoop

So, I have a task where job1 creates say file1, then job2 creates file2, then job1 creates file3. Right now, I have a variable i, which I increment in main, every time a job finishes. This works on my local machine, but when I run it on Amazon EMR, it doesn't. I believe that global counters can't be incremented in the main method? Is there a way I can solve this problem.