I have 10 indices in my elasticsearch cluster (4 nodes) and I want to run a term query on a field "x".
All of the 10 indices have the same exact mapping and the values of some fields might differ but the values of "x" are always the same across all the indices.
If I run a term query on the field "x", is it faster to randomly choose an index during runtime, or always run the query on one fixed index?
There may be multiple requests simultaneously.
I don't have resources to simulate the situation to experiment myself, hoping someone here can explain theoretically.