What is the query performance difference in querying to an index with all user data and querying to a particular index with half of the data?
For instance if i have two indexes male and female. How much query improvement will i get in comparison to querying over one index and then filtering out based on gender field?
If the second part of the query is something like a full text search and/or a geo sort by distance then having a filter to reduce the dataset will accelerate the execution.
So shall i make two indexes male and female instead of a single index and querying on gender='male' ? If so how much time will it save for some million user data.
I don't think you will really notice the difference.
But the only way to get an answer is to try it.
Injecting few millions random data is quite easy.
For example: https://github.com/dadoonet/legacy-search/tree/06-fuzziness?files=1
1 Like
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.