Hi Arize AX Support I have self hosted phoenix using docker. When I run a batch of 100 questions it is taking 35-40 mins to write the traces into Postgres and show it on UI. I have seen the cpu and memory usage of pods and GCP Postgres they are normal. I have enabled Prometheus to see where it is getting delayed from my observation phoenix batch insertions are taking time and if I try to scale the phoenix horizontally to enable multiple collectors I am seeing the database row locks exception in the logs. How do I improve the performance ? Please provide any inputs thank you.