I created a job which contains a step "sort rows" followed of a step " unique rows".
I run this job on 14 files which contain about 80000 records.
The execution lasts 7h30.
Is this normal? How can I improve the performance of my job.