I want to shuffle the data by key and write the data to sink.
I'm not use group by because of it is no need for me do any aggregate.
So I wonder whether flink sql has keyword like "distribute by" in spark.
I want to shuffle the data by key and write the data to sink.
I'm not use group by because of it is no need for me do any aggregate.
So I wonder whether flink sql has keyword like "distribute by" in spark.
It feels like a hack to me, but you could accomplish this with MATCH_RECOGNIZE and PARTITION BY. Hopefully there's a cleaner way that I'm not seeing.