Scroll api vs Search after api to query large time series data

I am using elastic search as time series db. I want to run queries on large data but I am confused about using scroll API or search after API. IMO scroll API is expensive for large data due to file descriptor handlers kept open. Scroll API also fetch all results in memory and return a result based on page size as we call scroll API. Can anyone guide me?

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.