Forum Discussion
Updating issue
TomMartens Greg_Deckler bcdobbs AlexisOlson
I have a datset existing on service with incremental refresh for nearly all the tables.
I often find myself needing to
1. Add new tables from new sources to that BI
2. Apply incremental refresh on the above sources
3. Edit/Delete Existing measure
4. Create new Measure
What is the most convenient way for me to deploy those changes to Prod without destroying existing partitons?
I currently do that with a DEV copy which does not have incremental configured initially. Once I bring on those changes to te DEV, I recreate incremental on all the tables and then upload, which is extremely difficult for a dataset containg many tables with incremental. It is extremely inconvenient.
Is there any tools exist out there/ some XMLA script on SSSMS that lets me do the above jobs on the PROD directly without going through the hassle everytime I have those reqs.
I tend to just connect direct to the dev model via xlma end point with tabular editor and add new tables there/configure new incremental when required. It's relatively straight forward in TE3 where you can right click to update table meta data once you've set it up. One done I use deployment pipeline to step through test and prod but you could connect straight to prod.
Would happily jump on a call and demo if you needed.
Hey smpa01 ,
you can do, this is exactly what ALM Toolkit was build for.
Of course you have to practice, a little.
Regards,
Tom
8 Replies
Hey smpa01 ,
if you are on Premium Per Capacity or Premium Per User you can use ALM Toolkit: Home Page - ALM Toolkit (alm-toolkit.com).
You can use ALM Toolkit to connect to the Power BI Desktop file and "synch" the changes to the dataset residing in Power BI Service without overwriting existing partitions 🙂
Regards,
Tom
TomMartens one follow-up Q to this. Does ALM toolkit allow to delpoy selective changes to the server?
Meaning, I have delpyed a partitioned dataset (PROD) on the server and my dev dataset (local) is not partitioned. I work on my dev copy to update/create new meaures/tables. Is it possible to select a handful of changes (new tables/ new measures/ edited measures) through ALM toolkit and deploy them on the partioned dataset on the server. I have not used ALM toolkit before, but I think it can be extremely useful for my case. Please forgive, if it sounds like a stupid question but I want to be extremely cautious before jumping on to something without understanding the implication first that might cost me the PROD.
bcdobbs I don't have TE3 yet
Hey smpa01 ,
you can do, this is exactly what ALM Toolkit was build for.
Of course you have to practice, a little.
Regards,
Tom
I believe this is possible but I haven't attempted it myself yet. XMLA connection is the key.
Here are some links that may be useful:
https://docs.tabulareditor.com/te3/powerbi-xmla.html
This is a documented limitation:
https://learn.microsoft.com/en-us/power-bi/create-reports/service-export-to-pbix- Reports created in the Power BI service that are based on a single dataset that's configured for large models or incremental refresh, or that has been modified by using the XMLA endpoint can only be downloaded in live connected report mode.
- Datasets modified by using the XMLA endpoint can't be downloaded to a .pbix file.
There are ways to reconstruct a .pbix but they're hacky:
https://data-goblins.com/power-bi/download-pbix
I tend to just connect direct to the dev model via xlma end point with tabular editor and add new tables there/configure new incremental when required. It's relatively straight forward in TE3 where you can right click to update table meta data once you've set it up. One done I use deployment pipeline to step through test and prod but you could connect straight to prod.
Would happily jump on a call and demo if you needed.
TomMartens AlexisOlson I am on Premium capacity, so it should not be an issue. I will give ALM toolkit and XMLA a spin and report back my experience on each
