Collecting file from local or HDFS is too slow.

[pppeace]

Context

I'm trying to collect corpus(above 200g, 3000w+ rows) to Doris

Phenomenon

I think 3~50rows/s is abnormal, whatever with localFile or HDFS.
The write/read count is close, so I guess the limit is on the read side.

Expect

Some advices or best practice. Or maybe my requirement is not a common scenario?

Others

And I just found an issue as same as mine.link, but no one help with this.😥

[pppeace]

anybody is here?

[Leeeee16]

I seem to be encountering the same issue, and I am using localfile, and mysql cdc to mysql, its too slow.

[EricJoy2048]

You need to describe the details:

seatunnel version:
jobconfig content:
submit job command: