[ 
https://issues.apache.org/jira/browse/STANBOL-471?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13194612#comment-13194612
 ] 

Ali Anil Sinaci commented on STANBOL-471:
-----------------------------------------

As the result of a discussion with @Rupert, following workflow seems convenient 
for a two layered store architecture:

S1) A store that simply stores the ContentItem with all content and metadata 
(enhancements)
S2) Semantic indexes created through LDProgramManager

1) User sends Content to the Contenthub
2) Content gets enhanced
3) The enhancement results are stored and indexed in [S1]
4) Contenthub indexes the ContentItem in all defined semantic indexes [S2]
4a) A default index (with common semantic fields such as the current 
"contenthub" index)
4b) A special user defined index with the name "myIndex"

Now assume that user submits 10.000 documents to Contenthub
Afterwards, assume that the user edits the LDPath program for "myIndex"

1) Contenthub creates a new SolrCore with the name "myIndex-1"
… in the meantime, the user can still use myIndex with the old schema
2) Contenthub starts to reindex the 10.000 enhanced documents by
2a) reading the Enhancements form [S1]
2b) indexing them by the new LDPath program for "myIndex" to the SolrCore 
myIndex-1
3) After the reindexing completes
3a) Contenthub replaces the SolrCore "myindex" with "myIndex-1"
3b) Contenhub shuts down the old "myindex" SolrCore and deletes the old data on 
the disc
                
> Two layered storage for Contenthub store
> ----------------------------------------
>
>                 Key: STANBOL-471
>                 URL: https://issues.apache.org/jira/browse/STANBOL-471
>             Project: Stanbol
>          Issue Type: New Feature
>          Components: Content Hub
>            Reporter: Ali Anil Sinaci
>
> Contenthub needs a two layered architecture to store the content items. For 
> the first layer, Contenthub should provide very basic indexing facilities for 
> legacy CMSs. On top of that, semantic indexed should be provided through 
> LDPath programs to index specific parts of the actual content and/or parts of 
> enhancements.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira


Reply via email to