[ 
https://issues.apache.org/jira/browse/OAK-1356?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13886457#comment-13886457
 ] 

Thomas Mueller commented on OAK-1356:
-------------------------------------

> if the capability of a repository to have an infinite transient space was 
> defined by JCR

I think I repeat myself, but the MongoMK does have that capability. The problem 
is performance: large transactions are very slow when the storage is remote. 
This is not something we can solve within the MongoMK, as the problem is really 
within oak-core (multiple scans are made over the whole transaction). The 
problem shows up with the MongoMK much more than with the TarNodeStore because 
the TarNodeStore _usually_ stores the changes locally, so that reading is fast. 
The problem in oak-core is something we probably should solve, that's true.

> we (or users) end up checking for files or other non-documented conditions

I suggest to write the application such that it doesn't rely on features that 
are not working well. And large transactions currently are not working well. 
Therefore, there is no need to check anything.

> Expose the preferred transient space size as repository descriptor 
> -------------------------------------------------------------------
>
>                 Key: OAK-1356
>                 URL: https://issues.apache.org/jira/browse/OAK-1356
>             Project: Jackrabbit Oak
>          Issue Type: Improvement
>            Reporter: Tobias Bocanegra
>
> The problem is that the different stores have different transient space 
> characteristics. for example the MongoMK is very slow when handling large 
> saves.
> suggest to expose a repository descriptor that can be used to estimate the 
> preferred transient space, for example when importing content.
> so either a boolean like: 
>   {{option.infinite.transientspace}}
> or a number like:
>   {{option.transientspace.preferred.size}}
> the later would denote the average number of modified node states that should 
> be put in the transient space before the persistence starts to degrade.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Reply via email to