I have a huge table (in Hadoop) partitioned by day.
From that table, I would like to create an anomaly table that will compare the locations of each day with the locations of the previous month.
If a location in the current day wasn't in the previous month it will add it in the anomaly table.
- I don't want to do it in code but only in SQL.
I know it should be possible in standard databases,
the question is: Is it possible in Hadoop and what will be the best way?
Thx