Best Option for Power BI Dashboard - Performance

[mhmmdrn]

Hi Community,

we dont have much experience with Power BI but we would like to create a couple of dashboards for our customers.
We have started as data storage with Azure Blob Container and by using Logic Apps, a csv file is generated and created in Blob for each type of dashboard (It means we have blob containers for each type of dashboard and customers).
I am trying to explain our scenario with a few bullets in order to make it clear. 

1. Customer count : 30

2. Dashboard Count: 8

3. Each csv data is apprx. 50 KB

Our main concern is the performance, if we use a lot of Blob Container connection with Power BI (with this example we would have 8*30=240 Blob Container Connection). For example, does it take long to load all the data every time when you open the dashboard?

Second option is Azure Table Storage but at this time it is nearly impossible to load all data into Table because you must to write every single row one by one with logic apps.

For another options we dont have enough experience. i.e SQL Server, SQL Service or other Data Storages in azure.

For any kind of suggestion related to our scenario above i would appreciate. Thanks.

[v-jianpeng-msft · Accepted solution]

Hi, @mhmmdrn 

You use Azure Blob Container to store your data. As far as I know, blob container performance depends on the following aspects: concurrency, request size, performance tier, upload source speed, and download destination. You need to know the request limit for a single blob, as shown in the following table:

A single blod can accept up to 500 requests per second, and if the number of requests sent by your dashboard exceeds this limit, you may report an error such as 500 (operation timeout) or 503 (server busy). In your scenario, this limit should not be reached.

Also, you are worried about whether you will have to wait for a long time every time you open the dashboard. You can use CDN access to blobs to enable the query caching feature in Power BI Service.

The reading of data in a blob is described as follows:

In addition, you need to select the appropriate blob based on the size of your data, and there is a different upper limit for different blob types.

We recommend that you use the Performance Check feature in Power BI desktop to check which of your queries are taking time.

You can refer to the article below to optimize your blob access:

Create a VM and storage account for a scalable application in Azure | Microsoft Learn

At the same time, you may want to understand the metrics of each blob to help you take it a step further in optimizing performance:

Performance and scalability checklist for Blob storage - Azure Storage | Microsoft Learn

How to Get Your Question Answered Quickly

Best Regards

Jianpeng Li

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