[
https://issues.apache.org/jira/browse/SQOOP-1602?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14258330#comment-14258330
]
Veena Basavaraj commented on SQOOP-1602:
----------------------------------------
[~stanleyxu2005] I am not sure I understand your last comment, when you say N
loaders, you mean there are N reducers ?
And what is Foo loader? is it one of the N loaders? This is important for us to
understand and one of my question in
https://issues.apache.org/jira/browse/SQOOP-1938 to understand why is the
reader/ writer sync requirement here.
What is your response to this comment earlier
{quote}
Loaders are currently executed on the same nodes as extractors (we're not doing
any re-shuffling of data), so the fact that we have unbalanced number of
loaders, means that we also had unbalanced number of extractors and that is the
real problem.
{quote}
> Sqoop2: Fix the current balancing across Loaders internal to Sqoop
> --------------------------------------------------------------------
>
> Key: SQOOP-1602
> URL: https://issues.apache.org/jira/browse/SQOOP-1602
> Project: Sqoop
> Issue Type: Improvement
> Reporter: Veena Basavaraj
> Assignee: Qian Xu
> Fix For: 1.99.5
>
>
> The balancing of the record to the loaders in done internally in SQOOP today
> While writing the Kite Connector Qian noticed that this is not done fairly.
> While I am testing kite connector, I allocated 2 loaders. I thought data will
> be divided by 50% and 50% to both loaders. But actually the second loader
> does nothing, because its DataReader does not have any data to provide. Is it
> by design?
> >> About loaders do not have data in a balanced way.
> My scenario is 4 "jdbc_mysql" extractors to extract 100k row data (10MB).
> There are 2 Kite loaders to read data.
> This must be a bug that needs to be fixed in SQOOP ( [~abec] confirmed it.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
