[
https://issues.apache.org/jira/browse/FLINK-16444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17286910#comment-17286910
]
Yu Li commented on FLINK-16444:
-------------------------------
After some more investigation, the metrics we plan to add here should compare
with [Kafka state store
metrics|https://kafka.apache.org/documentation/#kafka_streams_store_monitoring]
instead of [statistics and properties based
metrics|https://kafka.apache.org/documentation/#kafka_streams_rocksdb_monitoring].
I did some comparison between properties-based metrics listed in KIP-607 and
[those exist in
Flink|https://ci.apache.org/projects/flink/flink-docs-release-1.12/deployment/config.html#rocksdb-native-metrics],
and confirmed that {{live-sst-files-size}} is the only one we're missing and
may consider to add.
Regarding the [statistics-based
metrics|https://github.com/facebook/rocksdb/wiki/Statistics], it seems we don't
have them in Flink. It seems to be a little bit off-track to discuss here and
let's create some other JIRAs to follow up.
> Count the read/write/seek/next latency of RocksDB as metrics
> ------------------------------------------------------------
>
> Key: FLINK-16444
> URL: https://issues.apache.org/jira/browse/FLINK-16444
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / State Backends
> Reporter: Yun Tang
> Assignee: Yun Tang
> Priority: Major
> Fix For: 1.13.0
>
>
> Currently, user cannot know the read/write/seek/next latency of RocksDB, we
> could add these helpful metrics to know the overall state performance. To not
> affect the action performance much, we could introduce counter to only record
> the latency at interval of some actions.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
