Long query starving other small queries and slowing down the cluster


#1

Hi All,

I am on elastic search 5.4.

Problem -
I run 4 parallel queries and get following response times -
Q1 - 6234
Q2 - 12787
Q3 - 12881
Q4 - 16933

When I run sequentially -
Q1 - 4
Q2 - 6
Q3 - 761
Q4 - 16569

Setup
4 i3.4x nodes with 31G to ES and rest 30 G for FS and with 16 search threads per node.

As we can see, one long query is slowing down/starving rest of queries. Now my question is, is there a way if we can optionally fix search threads per query so that other small queries still get executed in justified time?


#2

Can anyone answer this?

Thanks


(Christian Dahlqvist) #3

No, as far as I know that is not possible.


(system) closed #4

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