[ 
https://issues.apache.org/jira/browse/IMPALA-13130?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18073521#comment-18073521
 ] 

ASF subversion and git services commented on IMPALA-13130:
----------------------------------------------------------

Commit 49ddba704f1eec308d276f9568e7564918677258 in impala's branch 
refs/heads/master from Michael Smith
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=49ddba704 ]

IMPALA-13164: Expose more KRPC metrics

Exposes Kudu RPC metrics that are already collected through the rpcz
JSON page: reactor_active_latency, reactor_load_percent, and
rpcs_timed_out_in_queue.

Reactor activity and load can help identify when reactor threads are a
bottleneck on RPC performance. Timed out RPCs could happen after
IMPALA-13130; a non-zero value would signify lost EndDataStream messages
reflecting significant problems in RPC flow in the environment.

Adds deserialization-queue-size and deserialization-queue-wait-time to
datastream-manager metrics to help identify when RPC deserialization is
slow and could benefit from increasing
datastream_service_num_deserialization_threads or the queue size.

Implements FunctionGauge to provide access to a thread-safe getter
providing the relevant metric.

Generated-by: Github Copilot (Claude Sonnet 4.6, GPT-5.3-Codex)
Change-Id: I62106bbab2b9290f0d5f456d4bb30005a5261524
Reviewed-on: http://gerrit.cloudera.org:8080/24179
Reviewed-by: Alexey Serbin <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>
Reviewed-by: Michael Smith <[email protected]>


> Under heavy load, Impala does not prioritize data stream operations
> -------------------------------------------------------------------
>
>                 Key: IMPALA-13130
>                 URL: https://issues.apache.org/jira/browse/IMPALA-13130
>             Project: IMPALA
>          Issue Type: Bug
>            Reporter: Michael Smith
>            Assignee: Michael Smith
>            Priority: Critical
>             Fix For: Impala 4.5.0, Impala 4.4.1
>
>
> Under heavy load - where Impala reaches max memory for the DataStreamService 
> and applies backpressure via 
> https://github.com/apache/impala/blob/4.4.0/be/src/rpc/impala-service-pool.cc#L191-L199
>  - DataStreamService does not differentiate between types of requests and may 
> reject requests that could help reduce load.
> The DataStreamService deals with TransmitData, PublishFilter, UpdateFilter, 
> UpdateFilterFromRemote, and EndDataStream. It seems like we should prioritize 
> completing EndDataStream, especially under heavy load, to complete work and 
> release resources more quickly.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to