I have to upload some terabytes of data that is text files.
What would be good option to do so:
i) using hadoop fs -put to copy text files directly on hdfs.
ii) copying text files as sequence files on hdfs ? What would be extra time
in this case as opposed to (i).