Optimizing Data & Queries For Global Use

Hey Team,

I wanted to see what the best practices are on this. I've been trying to do some testing, but it's inconclusive at best.

I own an SQL dataset with about 13 million rows * 60 columns, mix of dates, strings, numbers. It's about 7GB total. The data is metrics information for individual faclilites across the globe and the SQL table is being constantly fed new data, about 3 million rows a year across some ~50 facilities. The dataset is housed in a server in the US, but the data is accessed by users in those global plants.

Individual users only care about data as it pertains to their facility mostly, with some leaders wanting to see regional data, and some resources need access to the whole dataset.

Issue I'm encountiner is that the dashboards are chronically slow, from my understanding even if you have the visual slicers set to just EU, it'll still query the whole globe's data. Our facilities have terribly slow internet, sub 10MB/s for a whole facility.

What would be the best way to get user's the data that they are requesting in the most efficient manner? 

I'm trying to create & maintain one consistent App that can be used by all users globally. Currently its big DirectQuery Dataset that holds all the data, I'd like to not end up creating a dataset + Dashboard combo for every region. Is there some sort way for slicers to affect query's themselves? Does Import work better for these sort of situations? Has anyone encountered these sort of issues?

Thanks a ton,

David