[
https://issues.apache.org/jira/browse/IMPALA-11400?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Michael Smith updated IMPALA-11400:
-----------------------------------
Fix Version/s: (was: Impala 4.3.0)
> Kudu scan bottleneck due to sharing a single Kudu client for multiple tablet
> scans
> ----------------------------------------------------------------------------------
>
> Key: IMPALA-11400
> URL: https://issues.apache.org/jira/browse/IMPALA-11400
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Affects Versions: Impala 4.1.0
> Reporter: Sameera Wijerathne
> Priority: Major
> Labels: performance
> Attachments: 0.JPG, 1.JPG, 2-1.jpeg, 2.JPG, 2.jpeg, 3.JPG, 4.JPG,
> 5.JPG, Impala_1.png, Impala_2.png, Kudu_1.png, Kudu_2.png, WhatsApp Image
> 2022-06-07 at 10.39.27 PM.jpeg
>
>
> This issue was observed when impala queries large datasets resides in Kudu.
> Even single ImpalaD is scanning multiple kudu tablets, it shows a slowness to
> retrive data eventhough ImpalaD makes parrellel scans. Reason for this is
> ImpalaD only uses a single Kudu client for multiple scans but
> KuduScanner::NextBatch runs on a single thread. So it's rpc reactor thread
> utilizes upto a single core and bottlenecks all parrelel scans.
> This behaviour makes Impala clusters that scans kudu cannot be vertically
> scales to the maximum performance/cores of a node.
> Please refer the screenshots from Kudu slack channel for more information.
>
> !2-1.jpeg|width=717,height=961!
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
