[
https://issues.apache.org/jira/browse/HDFS-744?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13017598#comment-13017598
]
Hairong Kuang commented on HDFS-744:
------------------------------------
@Linyin, I am ok if we simply sync every packet to disk if the sync is set as a
first pass. This should work when sync calls are very frequent. If this turns
out to be a performance problem for hbase, we could file a separate jira to
optimize it by syncing to disk only when flush is called or at block boundary.
@Nicholas, what your suggested seems good to me. Could we do it in a separate
jira?
> Support hsync in HDFS
> ---------------------
>
> Key: HDFS-744
> URL: https://issues.apache.org/jira/browse/HDFS-744
> Project: Hadoop HDFS
> Issue Type: New Feature
> Reporter: Hairong Kuang
>
> HDFS-731 implements hsync by default as hflush. As descriibed in HADOOP-6313,
> the real expected semantics should be "flushes out to all replicas and all
> replicas have done posix fsync equivalent - ie the OS has flushed it to the
> disk device (but the disk may have it in its cache)." This jira aims to
> implement the expected behaviour.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
