Went through two telephonic conversation. In the second telephonic I got rejected because of negative feedback from interviewer however the answers which I gave was correct as per the core Hadoop. I am working on hortonworks distribution hence my answers was as per my experience with the Hadoop and core hadoop. MS is using cloudera distribution and I think that is the reason interviewer asked me questions on platform basis not technology basis like. How to split the table while Sqooping? My answer was, --split-by and here he was expecting answer like split-by-hash and split-by-mod which is completely cloudera platform dependent. However I had told interviewer that I am working on hortonworks distribution but he didn't considered that. Again he asked me about parquet and which is purely cloudera promoted file system.