[jira] [Commented] (KYLIN-2847) avoid doing useless work by checking query deadline

["Dayue Gao (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/KYLIN-2847?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16156762#comment-16156762 ] 

Dayue Gao commented on KYLIN-2847:
----------------------------------

[~mahongbin] could you take a look and review the patch?

> avoid doing useless work by checking query deadline
> ---------------------------------------------------
>
>                 Key: KYLIN-2847
>                 URL: https://issues.apache.org/jira/browse/KYLIN-2847
>             Project: Kylin
>          Issue Type: Improvement
>          Components: Query Engine, Storage - HBase
>    Affects Versions: v2.1.0
>            Reporter: Dayue Gao
>            Assignee: Dayue Gao
>         Attachments: KYLIN-2847.patch
>
>
> Under high load, requests can spend a long time waiting in RPC queue, probably longer than query timeout. However, current coprocessor timeout mechanism doesn't take RPC queue time into account. As a result, handling these requests not only waste server resources while doing nothing useful, but can also cause cascading failure when server crashes and client retries.
> To recover from server overload ASAP, we should check query deadline at each stages of query processing, and avoid spending resources on query that will exceed their deadline.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)