You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@kafka.apache.org by "Jorge Esteban Quilcate Otoya (Jira)" <ji...@apache.org> on 2022/12/04 23:54:00 UTC
[jira] [Created] (KAFKA-14441) Benchmark performance impact of metrics library
Jorge Esteban Quilcate Otoya created KAFKA-14441:
----------------------------------------------------
Summary: Benchmark performance impact of metrics library
Key: KAFKA-14441
URL: https://issues.apache.org/jira/browse/KAFKA-14441
Project: Kafka
Issue Type: Task
Components: metrics
Reporter: Jorge Esteban Quilcate Otoya
While discussing KIP-864 ([https://lists.apache.org/thread/k6rh2mr7pg94935fgpqw8b5fj308f2n7]) there is a concern on how much impact there is on sampling metric values, particularly when adding metrics that record values per-record instead of per-batch.
By adding benchmarks for sampling values, there will be more confidence whether to design metrics to be exposed at a DEBUG or INFO level depending on their impact.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)