Hi,
We are running a POC to see if we can use Paimon over S3 for a very high
throughput and low latency use-case. We have two flink jobs, one reads from
Kafka source, converts the data and writes to a Paimon/S3 table and the
other reads from Paimon/S3 table and writes to a Kafka sink. We are seeing
a very large number of HEAD and LIST requests per second in S3. These
requests are contributing heavily towards the AWS S3 costs and we are
probably being throttled as well since we are seeing backpressure on the
Kafka source. Any thoughts on what we can do to reduce the number of
head/list requests to S3? Any configuration settings that we could use that
will help in our situation? I would really appreciate any help you can
provide.

Thanks!

Reply via email to