A customer is complaining of poor performance from the job shown in the exhibit, which
aggregates and joins on ZIP code. Currently all partitioning is set to “Auto” and sort/partition
insertion is allowed.
What would give the best performance improvement for the job?

A.
Hash partition and sort on ZIP code on the input links to the Aggregator and Join stages.

B.
Hash partition and sort on ZIP code prior to the Copy stage,and use same partitioning on the
Aggregator and Join stages.

C.
Use same partitioning on all input links in the job; create the input data set by hash-partitioning
its input on ZIP code.

D.
Hash partition and sort on ZIP code prior to the Copy stage and use entire partitioning on the
output of the Aggregator stage.

Explanation: