[jira] [Commented] (UIMA-4685) Support different CAS serializations in CAS Editor

["Richard Eckart de Castilho (JIRA)" <de...@uima.apache.org>]

    [ https://issues.apache.org/jira/browse/UIMA-4685?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15074125#comment-15074125 ] 

Richard Eckart de Castilho commented on UIMA-4685:
--------------------------------------------------

I would guess this gets a lot faster if `-Duima.allow_duplicate_add_to_indexes` ( ALLOW_DUP_ADD_TO_INDEXES ) is specified (cf. FSIndexRepositoryImpl line 98ff and FSIntArrayIndex line 101ff). If dups are not allowed and there are many equal-but-not-identical FSes in the index, then the insert operation appears to get very slow. For simple annotations that would e.g. be the case if there are many annotations of the same type at the same offsets.


> Support different CAS serializations in CAS Editor
> --------------------------------------------------
>
>                 Key: UIMA-4685
>                 URL: https://issues.apache.org/jira/browse/UIMA-4685
>             Project: UIMA
>          Issue Type: New Feature
>          Components: CasEditor
>    Affects Versions: 2.8.1SDK
>            Reporter: Peter Klügl
>         Attachments: caseditor.csv, caseditor_bincas.csv
>
>
> Support different CAS serializations in CAS Editor. Some recent profiling revealed that, when opening a 35MB xmiCAS, the CAS Editor spends about 70% of its time for deserializing.



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