[jira] [Commented] (FLINK-16444) Count the read/write/seek/next latency of RocksDB as metrics

["Peidian Li (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-16444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17080283#comment-17080283 ] 

Peidian Li commented on FLINK-16444:
------------------------------------

I think we can expose more rocksdb metrics, such as:I think we can expose more rocksdb metrics, such as:block-cache-hit-count、block-read-count、block-read-byte、block-read-time、get-from-memtable-time、get-from-output-files-time、seek-on-memtable-time next-on-memtable-time、seek-child-seek-time、seek-child-seek-count、write-memtable-time、write-delay-time.These metrics are very useful for getting the performance of rocksdb.See also:https://github.com/facebook/rocksdb/wiki/Perf-Context-and-IO-Stats-Contexthttps://github.com/dataArtisans/frocksdb/blob/FRocksDB-5.17.2/include/rocksdb/perf_context.hhttps://github.com/dataArtisans/frocksdb/blob/FRocksDB-5.17.2/db/internal_stats.h

> 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: 1.11.0
>
>
> 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)