|
|
I need a text type 4 long text! At 2016-12-16 10:24:35, "JIRA [hidden email] [via Apache CarbonData Mailing List archive]" < [hidden email]> wrote:
[ https://issues.apache.org/jira/browse/CARBONDATA-433?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Jihong MA updated CARBONDATA-433:
---------------------------------
Description:
Currently, the blocklet size is configed through carbon.blocklet.size in carbon.properties file, its default value is 120000 (number of rows) across all tables. this can potentially cause compatibility issue when changing the setting.
we should consider to incorporate this information at file level, ideally stored in file footer, Carbon reader need to respect to the setting while reading.
was:
Currently, the blocklet size is configed with carbon.blocklet.size in carbon.properties file, its default value is 120000. This will cause we can not read the previous table if we change the blocklet size
Summary: Respect to blocklet size setting when scanning Carbondata files (was: blocklet size should belong to the table's meta data)
> Respect to blocklet size setting when scanning Carbondata files
> ---------------------------------------------------------------
>
> Key: CARBONDATA-433
> URL: https://issues.apache.org/jira/browse/CARBONDATA-433> Project: CarbonData
> Issue Type: Sub-task
> Reporter: suo tong
> Assignee: cen yuhai
>
> Currently, the blocklet size is configed through carbon.blocklet.size in carbon.properties file, its default value is 120000 (number of rows) across all tables. this can potentially cause compatibility issue when changing the setting.
> we should consider to incorporate this information at file level, ideally stored in file footer, Carbon reader need to respect to the setting while reading.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
To start a new topic under Apache CarbonData JIRA issues, email ml-node+ [hidden email]
To unsubscribe from Apache CarbonData Mailing List archive, click here.
NAML
|
Locked
