Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!

Reply
Anonymous
Not applicable

Live connection VS "import data" on premium via enterprise gateway performance

I was hoping to get some advice on which would yield fastest front-end dashboard performance out of the possible options below.

I'm aware of the other limitations of these methods, but looking to understand just the responsiveness of things like, applying several dashboard filters and displaying the row level data in a table.

 

Option 1: Import data from on premise SQL DB via gateway into shared PBI Premium capacity (incremental refresh)

Option 2: Live Query via well constructed OLAP/SSAS cubes via gateway into PBI Premium

 

My concern is that the project requires a lot of table visuals displaying row level searches in the dash, with multiple filters, on a large transactional dataset.

 

In Reza Rad's book "Pro Power Bi Architecture" he says:

"Import Data has the fully functional Power BI with a great performance.

So If your data set is not a huge dataset, then you can easily use the Import

Data method and produce reports in a very fast time frame."

 

My understanding is that via the PBI service somewhat auto-magically handles the underlying query optimisations & aggregations to pre-cache all potential views. However I'm thinking this could be problematic for large transactional data shown in row level table form in the dash?
Thankfully we've got premium capacity, so i know we can do up to 10gb data model, but i'm wondering if LiveQuery may provide better peformance? Or an idea of the point at which the dataset becomes so huge it's in-advisable to continue using import?

I realise this would depend heavily on the server resources available, but wondering in a scenario where the servers are comparably powered, which would more efficiently use these resources?

Postgre DB + SSAS --> Gateway --> PBI Premium Service (via live connection)

Postgre DB --> Gateway --> PBI Premium Service (via import)

 

I'm thinking LiveQuery may be better because rather than pre-caching all possible options, it's only going to pull what is needed "on demand" with only the handful of filter combinations needed by the user at the front end? Rather than trying to pre-cache & optimise queries for all possible views?
This may be a critical performance boost when trying to display row level data in table form in the dash?

 

Any advice is very much appreciated!

1 REPLY 1
v-kkf-msft
Community Support
Community Support

Hi @Anonymous ,

 

Do you have any requirements for the update frequency of the dataset?

 

If you don't need real-time data, and the size of the dataset does not exceed the limit in Premium capacity, then I think the Import mode will respond to queries faster. When the visuals are loaded, then they will query imported data and all changes to visuals will be reflected quickly.


If you need real-time data, try to use the Live Connection mode. When you interact with the visuals in dashboard, they will send the query to the data source and load the latest data.

 

Microsoft Power BI connectivity modes 

Power BI – Live connection vs. Import, comparison and limitations  

 


If the problem is still not resolved, please provide detailed error information or the expected result you expect. Let me know immediately, looking forward to your reply.

Best Regards,
Winniz

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

Helpful resources

Announcements
April AMA free

Microsoft Fabric AMA Livestream

Join us Tuesday, April 09, 9:00 – 10:00 AM PST for a live, expert-led Q&A session on all things Microsoft Fabric!

March Fabric Community Update

Fabric Community Update - March 2024

Find out what's new and trending in the Fabric Community.

Top Solution Authors
Top Kudoed Authors