[
https://issues.apache.org/jira/browse/HBASE-2467?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12859111#action_12859111
]
Todd Lipcon commented on HBASE-2467:
------------------------------------
Writing is still synchronized, but flushing isn't. Basically, each packet that
the DFS Client sends out gets a sequence number. Flush has two parts: (a) any
data that hasnt' been sent yet gets sent, and (b) client needs to wait for an
ack that this data has been received by the full pipeline. We used to do both
parts synchronized. HDFS-895 makes it so only part A is synchronized, and then
the waiting can happen concurrently with more writes, flushes, etc.
> Concurrent flushers in HLog sync using HDFS-895
> -----------------------------------------------
>
> Key: HBASE-2467
> URL: https://issues.apache.org/jira/browse/HBASE-2467
> Project: Hadoop HBase
> Issue Type: Improvement
> Components: regionserver
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Attachments: hbase-2467.txt
>
>
> HDFS-895 changes hflush() to be able to run concurrently from multiple
> threads, where flushes can be concurrent with further writes to the same file.
> We need to rip out/amend the group commit code a bit to take advantage of
> this.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
