To use multiple mappers in Sqoop, RDBMS table must have one primary key column (if present) in a table and the same will be used as split-by column in Sqoop process. If primary key is not present, we need to provide any unique key column or set of columns to form unique values and these should be provided to -split-by column argument.
What is the criteria for specifying parallel copying in Sqoop with multiple parallel map tasks?
To use multiple mappers in Sqoop, RDBMS table must have one primary key column (if present) in a table and the same will be used as split-by column in Sqoop process. If primary key is not present, we need to provide any unique key column or set of columns to form unique values and these should be provided to -split-by column argument.
