Don't miss your chance to take exam DP-600 or DP-700 on us!
Request nowLearn from the best! Meet the four finalists headed to the FINALS of the Power BI Dataviz World Championships! Register now
Hi, i got a glimpse of power automate today from one of my peers. I wondered immediatly if it is an alterantive to ssis for our daily ingestion of a few million records into our sql warehouse. Some of our current ssis etl sources include netsuite, oracle, sap, db2 and sql. Also, with adf and fabric also as alternatives why wouldnt i go in those directions? My limited understanding of adf is that its more of a big data tool and i dont want to call ssis from adf. My understanding of fabric is that its the thing that allows you to create flows (data, maybe work) but i've never heard of its flows ending in sql tables, just datasets. speaking of datsets, do they use the vertipac engine too just liks ssas and semantic models? I have to admit that i dont underatnd from the quick glimpse i got today the difference between power automate desktop and power automate cloud. but it sounded like desktop is more the details while cloud is more the scheduling.
Solved! Go to Solution.
Hi @db042190 ,
My first gut reaction is to say that no it is not. Power Automate is designed for workflows, approvals and event-driven automation. It can move small data sets around but as soon as you get to any scale, it can timeout, it can be throttled, and it generally needs quite a bit of care and feeding due to data exceptions or unexcpted interruptions in your data flow.
So, it is not really an ETL. However, you mentioned Fabric. In the world of Fabric, you can have pipelines and use cloud ETL quite nicely. Azure Data Factory is quite good at this and you don't get into issues with scale causing it to break. There might be more licensing costs though depending on your throughput and usage.
Proud to be a Datanaut!
Private message me for consulting or training needs.
Hi @db042190 ,
My first gut reaction is to say that no it is not. Power Automate is designed for workflows, approvals and event-driven automation. It can move small data sets around but as soon as you get to any scale, it can timeout, it can be throttled, and it generally needs quite a bit of care and feeding due to data exceptions or unexcpted interruptions in your data flow.
So, it is not really an ETL. However, you mentioned Fabric. In the world of Fabric, you can have pipelines and use cloud ETL quite nicely. Azure Data Factory is quite good at this and you don't get into issues with scale causing it to break. There might be more licensing costs though depending on your throughput and usage.
Proud to be a Datanaut!
Private message me for consulting or training needs.
| User | Count |
|---|---|
| 20 | |
| 18 | |
| 11 | |
| 9 | |
| 7 |
| User | Count |
|---|---|
| 42 | |
| 38 | |
| 18 | |
| 17 | |
| 16 |