You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-issues@hadoop.apache.org by "Chao Shi (JIRA)" <ji...@apache.org> on 2012/09/02 13:31:07 UTC

[jira] [Commented] (HDFS-3885) QJM: optimize log sync when JN is lagging behind

    [ https://issues.apache.org/jira/browse/HDFS-3885?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13446918#comment-13446918 ] 

Chao Shi commented on HDFS-3885:
--------------------------------

A similar one to save network latency: sync logs to lagging node in a larger batch. I guess a batch of 512K or 1MB should be much efficient.

Note that this can also work for uncommitted transactions. Imagine this with 3 JNs:
Tx1 is committed by JN1 and JN2. QJM is writing Tx2. JN3 is lagging. So we have tx1 and tx2 in its queue. We can send them to JN3 in a batch.

To implement the above idea, it needs more changes to current code structure, which simply uses a single threaded executor as the queue.
                
> QJM: optimize log sync when JN is lagging behind
> ------------------------------------------------
>
>                 Key: HDFS-3885
>                 URL: https://issues.apache.org/jira/browse/HDFS-3885
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>    Affects Versions: QuorumJournalManager (HDFS-3077)
>            Reporter: Todd Lipcon
>
> This is a potential optimization that we can add to the JournalNode: when one of the nodes is lagging behind the others (eg because its local disk is slower or there was a network blip), it receives edits after they've been committed to a majority. It can tell this because the committed txid included in the request info is higher than the highest txid in the actual batch to be written. In this case, we know that this batch has already been fsynced to a quorum of nodes, so we can skip the fsync() on the laggy node, helping it to catch back up.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira