I'm looking to migrate 15 terrabytes of data into Hadoop and considereing FTP or SQOOP. Can anyone advise on the maximum volumes that SQOOP can handle as I've been told that its not normally used above 10Gb.
The main question is "What is the source of data?"
if it's RDBMS then sqoop and answer is Yes..You can leverage sqoop to load 15TB of data
If it's not RDBMS then you should look into NiFi or Flume or if you just want to load data into HDFS then webhdfs
Thanks. The migration is from Oracle so sounds like SQOOP will work fine.