Dataset Refresh Slow on Service

[Anonymous]

I have a dataset that contains 3 big tables and 10 small tables. 1 of the big table is directly pulling data from Oracle database, the other 2 tables are referenced by this 1 big table. I have also done a lot of Power Query editing, such as renamed every column. Settings in datasource of Oracle database is import, not DirectQuerry.

Now, the dataset refresh on service is triggered by restful Api with a Powershell script. It takes about 10 mins on average if it is the only dataset runs the refresh on the gateway, 16 mins if it runs the refresh with other dataset refresh simultaneously. My end client wants me to reduce the refresh time.

I installed my On-Premise gateway on a virtual server has 32GB ram and 8 cores.

Can you give me some other hints for making it faster?

[GilbertQ]

Hi there

It would appear that possibly your On-Premise server is running at max CPU and memory when both data refreshes are running.

You would have to see which data refresh is CPU intensive.
Some idea would be when the data is refreshed via Power Query this can be an expensive CPU exercise and if there are other datasets running in parallel this could affect the performance. I would have a look at the other datasets also.

If I could ask why they would want it to refresh faster, when you can only refresh 8 times a day, which is hourly on Power BI Pro?