You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Vladimir Ozerov (JIRA)" <ji...@apache.org> on 2017/10/23 07:45:00 UTC

[jira] [Updated] (IGNITE-6412) Bypass GridCacheMapEntry during bulk data load

     [ https://issues.apache.org/jira/browse/IGNITE-6412?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Vladimir Ozerov updated IGNITE-6412:
------------------------------------
    Labels: performance  (was: iep-1 performance)

> Bypass GridCacheMapEntry during bulk data load
> ----------------------------------------------
>
>                 Key: IGNITE-6412
>                 URL: https://issues.apache.org/jira/browse/IGNITE-6412
>             Project: Ignite
>          Issue Type: Task
>          Components: cache
>    Affects Versions: 2.1
>            Reporter: Vladimir Ozerov
>              Labels: performance
>
> Currently every cache update goes through famous {{GridCacheMapEntry}} entry and acquire a monitor on it. This is done for the sake of thread-safety. 
> When doing bulk data load we can try the following approach:
> 1) Restrict all cache operations except of target data load 
> 2) Bypass {{GridCacheMapEntry}} completely, as we guarantee that there are no conflicting updates
> 3) Allow accessing cache when load finished
> NB: May be the same approach can be used even for plain cache operations due to our striped-pool architecutre. This needs to be investigated further.



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