Posting is not a problem. I can divide it into chunks and process it. That is a different story. I am trying to ask how are you handling the scalability/distribution of records across the worker nodes? Or are you dependent on Spark to handle it?