You are viewing a plain text version of this content. The canonical link for it is here.
Posted to oak-issues@jackrabbit.apache.org by "Chetan Mehrotra (JIRA)" <ji...@apache.org> on 2016/01/20 14:34:39 UTC

[jira] [Resolved] (OAK-2891) Use more efficient approach to manage in memory map in LengthCachingDataStore

     [ https://issues.apache.org/jira/browse/OAK-2891?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Chetan Mehrotra resolved OAK-2891.
----------------------------------
       Resolution: Later
    Fix Version/s:     (was: 1.4)

So far it has not posed a problem and this feature is also rarely used. If need arises can reopen the issue later

> Use more efficient approach to manage in memory map in LengthCachingDataStore
> -----------------------------------------------------------------------------
>
>                 Key: OAK-2891
>                 URL: https://issues.apache.org/jira/browse/OAK-2891
>             Project: Jackrabbit Oak
>          Issue Type: Improvement
>          Components: upgrade
>            Reporter: Chetan Mehrotra
>            Priority: Minor
>
> LengthCachingDataStore introduced in OAK-2882 has an in memory map for keeping the mapping between blobId and length. This would pose issue when number of binaries are very large.
> Instead of in memory map we should use some off heap store like MVStore of MapDB



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)