Pagination at the datasource level

Can Grafana paginate at the datasource level instead of at the client level?

As in you limit the results you load from the datasource, and then run the request/query again with an offset when you need more results. Versus loading all the results of a request/query and then Grafana paginates them in your browser.

My specific issue is that I’m loading logs via the MySQL (hence this subcategory) datasource pointed at GreptimeDB. If I load time periods with more than a few thousand results, the queries take much too long to load, if they load at all. To fix this I’d like to use datasource side pagination.

I have done some searching on this, and I think the answer is no, but I want to double check.

Thanks in advance. :slight_smile:

Yes, if it is supported, e.g. Loki:

But MySQL datasource doesn’t have this kind of feature implemented.

Ah, so it is something that needs to be implemented per datasource. That makes sense.

Thanks!