Advanced ETL beats SQL Merge

I have this ET that takes that takes about an hour to an hour and a half per day to run. I thought I would be smart and use the SQL merge functionality. However, ETL comes in faster @18 records per seconds while SQL merge came in at @15 records per second. I might let run a few days just to have a couple more data point, but it looks like be using ETL beats the SQL

Part of the reason I got ETL to go this fast is to use two different writers and a staging table as the source. The staging table has already determined which records are new and which ones are updates. Then I do a conditional split and send the new rows to a writer that says just add all. Remember, these records have already been predetermined. Then let the other writer update existing records as needed.

For updates, the performance depends on how big the table is and whether you have indexes on "WHERE FIELDS"

The target table is several million rows in it (and growing). If you notice, most of the rows are inserts. None the less, Advanced ETL is more than 10%.


The other part is that I uploading to Azure from my laptop. Other data flows work great.

Down the road, I am probably going to have to increase the DTU's on my Azure ($$$).