uima-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jerry Cwiklik (JIRA)" <...@uima.apache.org>
Subject [jira] [Closed] (UIMA-4896) UIMA-DUCC: instrument JP to log more details on negative performance metrics
Date Thu, 14 Apr 2016 19:26:25 GMT

     [ https://issues.apache.org/jira/browse/UIMA-4896?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Jerry Cwiklik closed UIMA-4896.
-------------------------------
    Resolution: Fixed

Added additional logging when negative analysis time is detected

> UIMA-DUCC: instrument JP to log more details on negative performance metrics 
> -----------------------------------------------------------------------------
>
>                 Key: UIMA-4896
>                 URL: https://issues.apache.org/jira/browse/UIMA-4896
>             Project: UIMA
>          Issue Type: Bug
>          Components: DUCC
>            Reporter: Jerry Cwiklik
>            Assignee: Jerry Cwiklik
>             Fix For: 2.1.0-Ducc
>
>
> Detect negative value for analysis time in UimaProcessContainer.doProcess() and log more
details. Performance breakdown for one job showed negative analysis time. This was only observed
and reported once. The hypothesis is that the JP provided bad data to the JD. Need proof that
this is indeed the case. 
> The JP code computes per CAS analysis time by doing before AE.process() and after snapshots
and than computing delta essentially:
> after.analysisTime() - before.analysisTime. The uima-core analysisTime is cumulative
and increasing each time process is called. In theory the delta should always be > 0. 



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

Mime
View raw message