1) Information from table A
2) Information from table B for all rows in A
3) Information from table C for all the rows in B.
Which of the following approaches would be better performance-wise:
1) Use 3 DB2 stages each to extract the above 3 information chunks i.e use join in the SQL statement (1 SQL statement would extract data from 'A', 2nd SQL statement would join A and B and the third would join A and C)
Or
2) Extract records from the 3 tables and join them using Join Stage
?
![Rolling Eyes :roll:](./images/smilies/icon_rolleyes.gif)
Also in the first approach would it be better to extract the data in datasets and then process it or should the data from DB2 stage be used directly without generating intermediate datasets?
Regards,
Bakul