[
https://issues.apache.org/jira/browse/FLINK-2055?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15984813#comment-15984813
]
Niels Basjes commented on FLINK-2055:
-------------------------------------
Last week I implemented a custom sink to feed my special records into HBase.
I use the BufferedMutator to do buffering and avoid going to HBase too often.
A problem I ran into which I also with this implementation is that of the times
when there are only a few records.
In my scenario these are the events recorded from our internal test instance of
our website.
The problem is that with these buffering solutions they all wait until the
buffer is full.
So the 'last event of to day' of that test site will appear in HBase tomorrow
morning when the first employee starts using it again.
What I think is needed (both in my and in this case) is a maximum buffering
period.
This will allow me to configure that I want a 'flush to hbase' at least every
second.
How should this be implemented in the Flink context?
> Implement Streaming HBaseSink
> -----------------------------
>
> Key: FLINK-2055
> URL: https://issues.apache.org/jira/browse/FLINK-2055
> Project: Flink
> Issue Type: New Feature
> Components: Streaming Connectors
> Affects Versions: 0.9
> Reporter: Robert Metzger
> Assignee: Erli Ding
>
> As per :
> http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Write-Stream-to-HBase-td1300.html
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
