hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Todd Lipcon (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HDFS-895) Allow hflush/sync to occur in parallel with new writes to the file
Date Thu, 11 Nov 2010 21:46:18 GMT

    [ https://issues.apache.org/jira/browse/HDFS-895?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12931218#action_12931218
] 

Todd Lipcon commented on HDFS-895:
----------------------------------

bq. Could this patch go without this fix? I would prefer to have a different jira to improve
hflush without data. I think I filed a jira a while back.

I agree that that bit of code is hard to understand. It also "works fine" without the fix,
but it does trigger an assertion if assertions are enabled -- a sequence number will get "skipped".
So I would prefer to keep the fix in, and in the JIRA you mentioned (avoid creating a packet
in the first place for empty flush) we can hopefully get rid of the confusing code. Is that
alright?

> Allow hflush/sync to occur in parallel with new writes to the file
> ------------------------------------------------------------------
>
>                 Key: HDFS-895
>                 URL: https://issues.apache.org/jira/browse/HDFS-895
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: hdfs client
>    Affects Versions: 0.22.0
>            Reporter: dhruba borthakur
>            Assignee: Todd Lipcon
>             Fix For: 0.22.0
>
>         Attachments: 895-delta-for-review.txt, hdfs-895-0.20-append.txt, hdfs-895-20.txt,
hdfs-895-review.txt, hdfs-895-trunk.txt, hdfs-895.txt, hdfs-895.txt, hdfs-895.txt
>
>
> In the current trunk, the HDFS client methods writeChunk() and hflush./sync are syncronized.
This means that if a hflush/sync is in progress, an applicationn cannot write data to the
HDFS client buffer. This reduces the write throughput of the transaction log in HBase. 
> The hflush/sync should allow new writes to happen to the HDFS client even when a hflush/sync
is in progress. It can record the seqno of the message for which it should receice the ack,
indicate to the DataStream thread to star flushing those messages, exit the synchronized section
 and just wai for that ack to arrive.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message