You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@hive.apache.org by "Kishen Das (Jira)" <ji...@apache.org> on 2020/05/28 21:56:00 UTC

[jira] [Updated] (HIVE-22014) [CachedStore] Tear down locks in CachedStore

     [ https://issues.apache.org/jira/browse/HIVE-22014?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Kishen Das updated HIVE-22014:
------------------------------
    Summary: [CachedStore] Tear down locks in CachedStore  (was: Tear down locks in CachedStore)

> [CachedStore] Tear down locks in CachedStore
> --------------------------------------------
>
>                 Key: HIVE-22014
>                 URL: https://issues.apache.org/jira/browse/HIVE-22014
>             Project: Hive
>          Issue Type: Sub-task
>            Reporter: Daniel Dai
>            Priority: Major
>
> There's a lot of locks in CachedStore. After HIVE-21637, only notification log puller thread will update the cache. And when it process event, the first thing is to mark the entry invalid. The only exception may be TableWrapperSizeUpdater, but we can also make it synchronous (maybe run it once after every iteration of notification log puller). There should be no synchronization issue and we can tear down existing locks to simplify the code.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)