Thank you for your feedback, appreciate your valuable time, skills and know ledges. Do you agree if we have more than 5 millions rows in source table, we should start using a process with increment approach (insert/update).
Its not so much the number of rows but the time it takes to refresh the data and how frequently
if 5 million row takes 2 mins to refresh then not much point in doing incremental - unless your window to refresh is 1 min
but if 10k rows takes 30 mins to refresh then incremental is likely to be a better option
but in either case - if you do a once a week refresh and you have 10 hours to refresh it then 30 mins does not matter either - so as usual "it depends" on your requirements whether you do incremental or full
for example on one of my servers I copy 12 Million rows onto 7 different tables once a week (with biggest table containing 3 million rows) - takes 3 mins to copy everything sequentially - why would I bother with implementing a process to do incremental processing?