StarRocks, a Linux Foundation project, is a next-generation sub-second MPP OLAP database for full analytics scenarios, including multi-dimensional analytics, real-time analytics, and ad-hoc queries. StarRocks received InfoWorld’s 2023 BOSSIE Award for best open source software.

StarRocks

Hello, team.
I'm loading data from kafka through flink starrocks connector. 
However, only the leader fe node has a very high CPU and memory usage rate. 
I installed starrocks on eks and exposed the 9030 port through nlb and the fe proxy outside through alb. 
The settings of Flink(jdbc-url, load-url) are set to domain of nlb, alb
Is the issue I'm going through normal?

As far as I know, the streamload api load should be distributed to each fe nodes. I confirmed through the log that transaction api calls were evenly distributed to 3 actual fe proxy nodes.

Leader FE is expected to have higher resource consumption compared to the follower since it's responsible for writing metadata while the followers don't. If it's too high, it could be that the stream load batch writing has very high frequency, and possibly small in size