[jira] [Updated] (HUDI-6446) Defer Initialization of MDT just at the end of first commit

2023-06-26 Thread ASF GitHub Bot (Jira)


 [ 
https://issues.apache.org/jira/browse/HUDI-6446?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

ASF GitHub Bot updated HUDI-6446:
-
Labels: pull-request-available  (was: )

> Defer Initialization of MDT just at the end of first commit 
> 
>
> Key: HUDI-6446
> URL: https://issues.apache.org/jira/browse/HUDI-6446
> Project: Apache Hudi
>  Issue Type: Improvement
>  Components: metadata
>Reporter: sivabalan narayanan
>Priority: Major
>  Labels: pull-request-available
>
> For a fresh table, when both FILES and RLI is enabled, we use default values 
> for num file groups i.e 10 for RLI. and this also creates a log file and does 
> not create a base file since there are no records to instantiate as such. So, 
> we should defer the instantiation to later. either at the end of first commit 
> or when the data table has atleast 1 completed commit. 
> For an already existing table, this is not an issue since if there are valid 
> records, we will dynamically determine the number of file groups. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)


[jira] [Updated] (HUDI-6446) Defer Initialization of MDT just at the end of first commit

2023-06-26 Thread sivabalan narayanan (Jira)


 [ 
https://issues.apache.org/jira/browse/HUDI-6446?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

sivabalan narayanan updated HUDI-6446:
--
Epic Link: HUDI-466

> Defer Initialization of MDT just at the end of first commit 
> 
>
> Key: HUDI-6446
> URL: https://issues.apache.org/jira/browse/HUDI-6446
> Project: Apache Hudi
>  Issue Type: Improvement
>  Components: metadata
>Reporter: sivabalan narayanan
>Priority: Major
>
> For a fresh table, when both FILES and RLI is enabled, we use default values 
> for num file groups i.e 10 for RLI. and this also creates a log file and does 
> not create a base file since there are no records to instantiate as such. So, 
> we should defer the instantiation to later. either at the end of first commit 
> or when the data table has atleast 1 completed commit. 
> For an already existing table, this is not an issue since if there are valid 
> records, we will dynamically determine the number of file groups. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)