'limit()' function optimization suggestion

[sonjoosik]

When I have a query that requests millions of series, I sometimes want to see only several samples out of the series. According to the document, the limit() function is supposed to be used for that purpose. However, the implementation seems different from what I expected. I do not want to send such a large number of queries to Cassandra to get only several samples. Can we optimize the limit() function's implementation a little bit?

[sonjoosik]

By the way, thank you for the great tool. I'm using Graphene every day, and it is helping me reduce a lot of time to analyze the metrics of our system. 👍

[Dark0096]

I really want to thank you for your attention. :)

It's a new perspective and a very difficult problem.
Because there is a problem in the lifecycle part. Each function performs various processing after fetching time series data( ES & Cassandra ). In other words, Graphene cannot know how many data fetching to data source.

I think there are two ways to resolve this issue.

The first is that graphene receives related information as a query parameter like from, until, and limit. This method looks best to me, but there is the huddle of having to fix Grafana.

Alternatively, it is processed by utilizing parameters within a specific function( like Limit function) before executing the function. However, this part is not good for me.

I will consider this part a little more with @jerome89.

Thanks again for your interest. :)