You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@kafka.apache.org by "Sophie Blee-Goldman (Jira)" <ji...@apache.org> on 2020/06/16 23:45:00 UTC

[jira] [Created] (KAFKA-10177) Replace/improve/remove Percentiles metric

Sophie Blee-Goldman created KAFKA-10177:
-------------------------------------------

             Summary: Replace/improve/remove Percentiles metric
                 Key: KAFKA-10177
                 URL: https://issues.apache.org/jira/browse/KAFKA-10177
             Project: Kafka
          Issue Type: Improvement
          Components: metrics
            Reporter: Sophie Blee-Goldman


There's an existing – but unused – implementation of percentile metrics that we attempted to use for end-to-end latency metrics in Streams. Unfortunately a number of limitations became apparent, and we ultimately pulled the metrics from the 2.6 release pending further investigation/improvement.

The problems we encountered were
 # Need to set a static upper/lower limit for the values
 # For reasonable accuracy over a large range of possible values (see above), a lot of memory per metric was required.
 # Possible memory leak, which combined with issue #2 caused us to hit OOM and die

Since the Percentiles class is part of the public API, we may need to create a new class altogether and possibly deprecate/remove the old one. Alternatively we can consider just re-implementing the existing class from scratch, and just deprecating the current constructors and associated implementation (eg the constructor accepts a max)



--
This message was sent by Atlassian Jira
(v8.3.4#803005)