no wonder, try to increase the qps and see when th...
# troubleshootingx
Xiang Fu
10/28/2022, 9:04 AM
no wonder, try to increase the qps and see when this
schedulerWaitMsp
Priyank Bagrecha
10/28/2022, 9:19 AMi am confused. do you mean to say start at a lower qps and increase it to see how far i can push?
Priyank Bagrecha
10/28/2022, 9:19 AMi thought this was an over provisioned cluster.
x
Xiang Fu
10/28/2022, 10:19 AM
did you config the query thread?
Xiang Fu
10/28/2022, 10:19 AM
cc: @Jackie for more visibility
Xiang Fu
10/28/2022, 10:20 AM
Also make sure you remove this https://github.com/apache/pinot/blob/master/kubernetes/helm/pinot/values.yaml#L102
-XX:ActiveProcessorCount=2p
Priyank Bagrecha
10/28/2022, 10:58 AMwe don't have active processor count
x
Xiang Fu
10/28/2022, 11:02 AM
hmm, ok, then I would suggest to use replicaGroup
Xiang Fu
10/28/2022, 11:02 AM
how many segments you have
Xiang Fu
10/28/2022, 11:03 AM
Xiang Fu
10/28/2022, 11:03 AM
If you have many segments, then each query will fanout to all the servers, that has negative impact on the qps
p
Priyank Bagrecha
10/28/2022, 11:04 AM100 segments. total table size is 128 GB