Thanks everyone for the replies and feedback. I plan to split the Spark tiering feature in two phases: 1. Phase 1: Extract the common code into fluss-client
2. Phase 2: Add the Spark tiering functionality @yuxia I agree that the default configuration should be optimal and that users shouldn't need to worry about it in most cases. However, I believe users should have the flexibility to adjust parameters based on their specific deployment scenarios, since use cases can vary significantly across different environments.
