11 Comments

1. 

   smruti ranjan mishra on May 18, 2020 at 6:18 am

   Nice article. Helped me save time. 🙂

   Reply
2. 

   Robbin h on August 4, 2020 at 10:47 am

   And how to define triggers ;)?

   Reply
   • 

     Erwin on August 14, 2020 at 12:51 pm

     HI Robbin, my triggers are always named like Daily/Weekly/Monthly followed by the name of the Pipeline. This way you keep it clear. I never link multiple Pipelines to the same Trigger.

     Reply
3. 

   Koen on September 15, 2020 at 9:27 am

   I’m wondering, Erwin, why do you use abbreviations?

   I’ve found that modern systems don’t really have a limit on number of characters. In Ye Olden Days those limits enforced abbreviations, modern systems don’t have this limitation.
   I’ve also found that the context-sensitivity of abbreviations means they make reading and interpreting the names more difficult.

   Wouldn’t it make sense, in the interest of legibility, to use the full name of just about everything?

   What is the advantage of
   LS_ABLB
   over
   LinkedService_AzureBlobStorage
   ?

   I can see a point in abbreviating LinkedService to LS; I mean; it should be clear from the context that this is a linked service. But ABLB, to me, is a lot harder to read and interpret than AzureBlobStorage. The result is that doing maintenance will be more difficult on the former than on the latter.

   Reply
   • 

     Erwin on September 15, 2020 at 9:55 am

     Dear Koen,

     These naming conventions are more of a guideline. I use them in this way to at least ensure that all LinkedServices / DataSets / Pipelines are built in a consistent way, but also that you can validate them with Test scripts in Azure Dev Ops.
     The most important thing about this blog is that you do it the same everywhere, if you use a different name for this, that is of course no problem. Most abbreviations are easily translatable to the correct Azure Service, especially if you work with it on a daily basis.

     Reply
4. 

   Manuel on October 8, 2020 at 1:56 pm

   Excellent article!
   How do you suggest define folder structures with pipelines? I think two general areas: Staging and DWH and categorized them by project/DDBB..¿?

   Reply
   • 

     Erwin on October 8, 2020 at 4:34 pm

     Happy to hear that Manual. You suggestion is a good one.

     Currently I’m using, And then you can still create a new sub for project. But probably you have also shared recourses across you different projects, so then my advice is then to add a Shared Project folder

     01.Datalake
     01.Control
     02.Command
     03.Execute
     02.Deltalake
     01.Control
     02.Command
     03.Execute
     03.DataStore
     01.Control
     02.Command
     03.Execute
     Work In Progress

     Let me know you findings

     Reply
5. 

   Hennie on March 24, 2021 at 3:12 pm

   Very well written Erwin! Keep m coming. I miss snowflake 😉 LS_SNOW_?

   Reply
   • 

     Erwin on March 25, 2021 at 6:15 pm

     Good one Hennie, I need to take some to review all the new sources

     Reply
6. 

   Rohan on June 13, 2022 at 1:26 pm

   Hi Erwin, when we link a datalake to Synapse it automatically adds linked services for datalake with names as -WorkspaceDefaultStorage. How do we change it to our naming conventions?

   Reply
   • 

     Erwin on June 13, 2022 at 2:37 pm

     Hi,

     You cannot, you need to leave them as they are. There’s no option yet to override these values as they belong to the initial deployment.

     Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.