DRAG DROP
You are developing a SQL Server Integration Services (SSIS) package that loads data into a
data warehouse hosted on Windows Azure SQL Database.
You must combine two data sources together by using the ProductID column to provide
complete details for each record. The data retrieved from each data source is sorted in
ascending order by the ProductID column.
You need to develop a data flow that imports the data while meeting the requirements.
How should you develop the data flow? (To answer, drag the appropriate transformation
from the list of transformations to the correct location in the answer area.)
Answer: 
Merge not Merge join (data is already sorted)
The Differences Between Merge and Union all are
1-union all the data must not be sorted while in Merge and Merge Join it has to be sorted
2-in union all you can have 2 or more inputs while in merge and Merge Join must be two only
So here the data is sorted and only two inputs but we have to join them on the Product ID column
So Merge Join is the correct answer
Thanks for clearing things up. I was confused between Q.85 and that one.
I think it is Merge Join because it combines two table using Product ID
More than anything else, the reason above scenario requires a MERGE JOIN is found in the question: “…to provide complete details for each record.”
Assume TableTest1 has two columns and tableTest2 has three columns, the output in TableTest3 would have MORE details per each record-set: that is, five columns.
**Place two packs of playing cards side-by-side on a table:
Merge: stacks the ‘sorted’ packs one-whole-pack-on-top of the other
Merger Join: interleaves the ‘sorted’ pair one-card-at-a-time into the other
Union All: aggregates the two packs of card together without order (unsorted)
Not Exactly
Merge: Interleaves the ‘sorted’ records from the two sources into the output (includes all fields)
Merge Join: Performs (INNER, LEFT OUTER, FULL OUTER) join on the two sources where the join fields and output fields must be identified
“to provide complete details for each record”. Hence Merge Join is the answer.