November 6, 2023 at 4:57 pm
My apologies in advance, as this is probably a dumb question -- I figured I risk asking just the same.
In 2023 is still worthwhile to learn on-prem SSIS, and SSAS. I'm just getting started in Data Development, ETL, SQL development.
It looks like Microsoft has steered their recent training on Microsoft.learn to Azure related technologies. I'm just curious what the experts think.
Thanks!
November 6, 2023 at 5:04 pm
Not a dumb question at all. In my opinion, MS is not focusing at all on SSIS/SSAS.
If you have the choice of what to learn, Azure and DataFactory are the new tech in town and I'd suggest going down that avenue.
The absence of evidence is not evidence of absence
- Martin Rees
The absence of consumable DDL, sample data and desired results is, however, evidence of the absence of my response
- Phil Parkin
November 6, 2023 at 6:06 pm
To play Devil's advocate, I would say it really depends. If you are currently employed, what does your employer use? I know at my workplace, we are not using SSAS at all, but have a pretty strong SSIS footprint at the moment (it's going to mostly go away soon).
If you currently don't have employment, it really depends on who you are applying to work for and how strong their technical team is. Applying at a small business, they are likely to host things in the cloud as it is cheaper than on-premise. Medium to Large businesses are likely to have their own servers and MAY have some data they consider to be highly confidential and don't allow it to touch cloud servers, so SSIS may be in place. The area I live in is not very tech-heavy, so finding a full time technical job is challenging, but most places are more comfortable working with the "tried and true" approach where the cloud still feels like a fad and they like having the servers sitting in a room they can control. So the adoption of Azure and other cloud technology is pretty light and SSIS/SSAS is a huge asset.
That being said, I agree with Phil that the Azure approach is going to be better long term as more and more companies migrate more and more stuff to the cloud.
The above is all just my opinion on what you should do.
As with all advice you find on a random internet forum - you shouldn't blindly follow it. Always test on a test server to see if there is negative side effects before making changes to live!
I recommend you NEVER run "random code" you found online on any system you care about UNLESS you understand and can verify the code OR you don't care if the code trashes your system.
November 6, 2023 at 7:12 pm
I agree with both responses.
In my experience, once Microsoft stops investing in a product for a few versions it is most likely because they are working on something new. We saw that with SSRS and with SSAS Multidimensional. If I had to place a bet, I would bet that we will see an on-premises version of ADF in the not-too-distant future. Many companies are still on-premises, and that will not change very quickly due to regulations etc.
So if you have a choice I would also recommend starting with ADF, but SSIS is also not a bad place to start learning...especially if you can't afford to spin up an Azure instance and pay the costs associated with it. Nothing in the cloud is free.
November 6, 2023 at 9:23 pm
ADF feels like 'the next version of SSIS, but in the cloud', IMHO. SSIS/SSDT are free; if you have access to materials to help you learn the on-prem tools, then ADF would seem more like a skills upgrade than learning a new product. The big difference are the methods through which you run the thing once it's deployed, but those aren't deep concepts.
Although I've been fiddling with the various incarnations of this product starting with DTS on SQL 2000, I never took the time to actually learn it properly. I recently had a project on ADF and went from 'never heard of it' to deploying a daily ASF pipeline that shreds client text files, imports them, and processes the results into client tables without much poking around or pouring through documentation. I just treated it like SSDT in a browser.
Obviously, there are much more to these tools than I bothered to learn, but it seems like you're not harming your future by choosing one over the other to get started.
Also, and I encourage folks with more knowledge on this than me to add to/correct this, but there's likely plenty of opportunities turning on-prem SSIS packages into ADF pipelines, for which knowledge of both would be beneficial.
Eddie Wuerch
MCM: SQL
November 7, 2023 at 9:47 pm
If I had to place a bet, I would bet that we will see an on-premises version of ADF in the not-too-distant future. Many companies are still on-premises, and that will not change very quickly due to regulations etc.
.
Thanks all for the advice! Have any of you experimented with Azure Data Factory's Self-Hosted Runtime? I've not read through all the documentation yet, but sounds like on-prem azure Data factory.
November 7, 2023 at 9:54 pm
Have any of you experimented with Azure Data Factory's Self-Hosted Runtime? I've not read through all the documentation yet, but sounds like on-prem azure Data factory.
The Self-Hosted IR is really just a gateway for data sources internal to your network and not accessible to the ADF instance in the cloud. I wouldn't describe it as on-prem Data Factory, because the processes still run in the cloud.
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply