You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Yifan Cai (Jira)" <ji...@apache.org> on 2024/08/12 18:34:00 UTC

[jira] [Commented] (CASSANDRA-19827) [Analytics] Add job_ideal_timeout_seconds writer option

    [ https://issues.apache.org/jira/browse/CASSANDRA-19827?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17873029#comment-17873029 ] 

Yifan Cai commented on CASSANDRA-19827:
---------------------------------------

CI is green

> [Analytics] Add job_ideal_timeout_seconds writer option
> -------------------------------------------------------
>
>                 Key: CASSANDRA-19827
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-19827
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Analytics Library
>            Reporter: Yifan Cai
>            Assignee: Yifan Cai
>            Priority: Normal
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> Option to specify the ideal timeout in seconds for bulk write jobs.
> It is only effective when the bulk write job is using S3_COMPACT data transport mode.
> When JOB_IDEAL_TIMEOUT_SECONDS is specified and less than the actual time the bulk write job
> needs to achieve the specified consistency level, it is ignored and job only exit after the desired consistency level has been satisfied.
> For example, a bulk write job indeed requires 1 hour to achieve LOCAL_QUORUM, it ignores
> any JOB_IDEAL_TIMEOUT_SECONDS that is less than 3600 seconds (1 hour), and only complete after 1 hour.
> If JOB_IDEAL_TIMEOUT_SECONDS is 5400 seconds (1.5 hours), the job after achieve LOCAL_QUORUM waits for at most 0.5 hours in addition. The effective
> wait time is the minimum of the remaining time to ideal timeout and the estimated wait time to finish all slice import (as estimated
> in org.apache.cassandra.spark.bulkwriter.ImportCompletionCoordinator).
> The ideal timeout is ignored in order to complete the bulk write job in some circumstances, hence named "ideal".



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

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org