When dealing with huge datasets, it is quite impossible that the code successfully executes on your personal desktop. You either need a locally installed clustered environment i.e. Hadoop Map-Reduce or a Cloud such as AWS. Here's an example of running such Job on AWS cloud.