[
https://issues.apache.org/jira/browse/SYSTEMML-145?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Deron Eriksson reassigned SYSTEMML-145:
---------------------------------------
Assignee: Deron Eriksson
> Crc file created for each output file when run in standalone mode
> -----------------------------------------------------------------
>
> Key: SYSTEMML-145
> URL: https://issues.apache.org/jira/browse/SYSTEMML-145
> Project: SystemML
> Issue Type: Improvement
> Reporter: Deron Eriksson
> Assignee: Deron Eriksson
> Priority: Minor
>
> When a DML script is executed in standalone mode (e.g.,
> ./runStandaloneSystemML.sh algorithms/datagen/genLinearRegressionData.dml
> -nvargs numSamples=1000 numFeatures=50 maxFeatureValue=5 maxWeight=5
> addNoise=FALSE b=0 sparsity=1000 output=linRegData.csv format=csv), a crc
> file is created for each file.
> For example, with the above script, the following files were written to the
> file system:
> linRegData.csv
> .linRegData.csv.crc
> linRegData.csv.mtd
> .linRegData.csv.mtd.crc
> This litters the file system with a lot of basically unnecessary files. If
> several scripts are executed, the number of crc grows rapidly.
> There should be a mechanism for one of the following:
> 1) don't generate crc files in the first place
> 2) delete generated crc files
> 3) have an option to generate crc files, with option by default set to false
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
