tabular data model in data store lake for power BI

Question

i have requirement to consume data from data lake store and display in power BI. data size is big it may be 20 GB or 50 GB, i don't know what size will come in feature.also i cannot use Azure DW and Azure Analysis service.

now i proposed a solution i should create a tabular file(pivot table) in azure data lake store with aggregated date with all table joins in data like store

Geography Name Product-Name Salesyerar2015 Salesyerar2017 userId

i need to show two year of data in power BI report.

for month level i have 24 columns for measures. in power i use import data from data lake store.

if data size is more i suggest power BI Premium for bigger data file so it should accommodate in cache( 50 GB limitations)

Now for small data set it is work fine, i cannot able to test big files.

now i want to know when i have date level tabular file is required how i accommodate in tabular model.

also i want to know this proposed design is correct or do we have any other approach to handle this use case.

Regards, manish

Hi, Any body can Microsoft Suggest me, it is urgent to communicate Client. — user3843858
Hi all, it's really hard to me understand what is best way to connect Power BI and data lake store. we cannot say size of Data Lake Store= Size of PowerBI — user3843858

Paul Andrew Paul Andrew · Accepted Answer · 2017-07-21T08:37:45

I would recommend you take a look into the U-SQL sampling function available out of the box in the Azure Data Lake Analytics service to reduce your datasets to a size that can be consumed by Power BI. It's a technique we use a lot.

Example U-SQL:

@Output =
    SELECT
        *
    FROM 
        @ParsedJSON
    SAMPLE UNIFORM
        (0.04); //4%

Then output the files to data lake store for consumption.

Hope this helps

tabular data model in data store lake for power BI

1 Answers