Hello there,
I am working on a project where we are leveraging Kafka Streams for real time data analytics; and I am seeking some advice 
on optimization strategies.
Our setup involves processing a high volume of streaming data; and while we are achieving good throughput; we are encountering some performance bottlenecks that are impacting our processing latency.
Despite configuring appropriate number of threads and instances, we’re seeing latency spikes during peak hours.
Our RocksDB state stores are occasionally experiencing performance degradation. We have tried adjusting cache sizes; but it has not resolved the issue entirely.
CPU and memory utilization on our processing nodes seems suboptimal. We have scaled up resources; but we are still encountering limits.
I am interested in hearing from anyone who has experience optimizing Kafka Streams for high throughput scenarios.
Also, I have gone through this post; https://ask.lenses.io/t/unable-to-see-collect-aws-msk-metrics-with-lenses-ccsp/ which definitely helped me out a lot.
Best practices for configuring Kafka Streams and RocksDB for high performance…
Thanks in advance for your help and assistance. 