Scheduled refresh through On-prem data gateway takes more time for different users

[Anonymous]

Hi all,

There is an on-prem gateway that is given access to 9 people in my organisation including me.

When I scheduled refresh for a particular huge dataset, it is taking a very long time to refresh while that is not the case with other person who has access to this same gateway and is scheduling refresh for the very same dataset in a different workspace.

Mine is taking like 1+ hour extra compared to what is taking for the other person.

Can someone help in understanding what might be the reason behind this?

Point to Note :

The machine where this gateway is installed is far from me. Like it's a remote machine managed by some IT people.

If someone can provide an insight over this quickly, it will be much helpful.

Thanks in advance.

[TomMartens]

Hey @Anonymous ,

to provide guidance, it's necessary to clarify some aspects.

A dataset originates from publishing a Power BI report to a workspace in the Power BI Service. It's simply not possible that the SAME dataset is part of two different workspaces.

A data gateway connection configured using a specific gateway is not specific to a dataset or workspace. A data gateway connection can be used with different datasets in different workspaces.

Workspaces are tied to a capacity. This capacity is either a Shared capacity (meaning backed by Pro licensing) or a Premium capacity. If both workspaces are backed by Pro licensing, then the workspaces are located in the same region as your Power BI tenant. If this is the case for both workspaces, then it's a little weird, because the distance of data travel is the same.

If Premium capacities are involved than different distances of data travel can exist.

Maybe there is anohter reason, your dataset might leverage data from the same data source but also use different queries or additional data sources like data from files that are located in a "slower" place, like a SharePoint library.

Hopefully, this adds some additional information to tackle your challenge.

Regards,

Tom

[Anonymous]

Thanks Tom, the datasets are not exactly the same . But very similar. Like the backend tables used for this has same size & no.of rows & columns. Just two different copies.Also the report which caused this dataset is also very similar in modelling & structure.Or to be exact just a .pbix file which has same template as the other.