[
https://issues.apache.org/jira/browse/HBASE-5954?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14222294#comment-14222294
]
Lars Hofhansl commented on HBASE-5954:
--------------------------------------
I don't mind having a look. Still looking into how it fits best. Seems that the
options are:
# have each SyncRunner keep two sequences and sequence numbers (sync and fsync)
and work off a single queue
# have different SyncRunners for sync and fsync and have two sync queues as well
#1 will be easier but a bit messy
#2 bit cleaner potentially will make it to share work between threads
I have a half baked patch for #1.
As we approach a world with ubiquitous SSDs this will become important.
> Allow proper fsync support for HBase
> ------------------------------------
>
> Key: HBASE-5954
> URL: https://issues.apache.org/jira/browse/HBASE-5954
> Project: HBase
> Issue Type: Improvement
> Components: HFile, wal
> Reporter: Lars Hofhansl
> Assignee: Lars Hofhansl
> Priority: Critical
> Fix For: 2.0.0
>
> Attachments: 5954-trunk-hdfs-trunk-v2.txt,
> 5954-trunk-hdfs-trunk-v3.txt, 5954-trunk-hdfs-trunk-v4.txt,
> 5954-trunk-hdfs-trunk-v5.txt, 5954-trunk-hdfs-trunk-v6.txt,
> 5954-trunk-hdfs-trunk.txt, hbase-hdfs-744.txt
>
>
> At least get recommendation into 0.96 doc and some numbers running w/ this
> hdfs feature enabled.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
