hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Suresh Srinivas (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-1295) Improve namenode restart times by short-circuiting the first block reports from datanodes
Date Wed, 20 Apr 2011 23:09:06 GMT

    [ https://issues.apache.org/jira/browse/HDFS-1295?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13022414#comment-13022414
] 

Suresh Srinivas commented on HDFS-1295:
---------------------------------------

Comments:
# Minor: TestDatanodeBlockScanner - could you LOG.info or LOG.debug instead of System.out
# Is it worth retaining the printDatanodeAssignments() and printDatanodeBlockReports(), which
probably was added as debug code?
# In the test we have TIMEOUT to be 20s. Is it reasonably long enough so that tests do not
fail?
# In block report time, why is the report creation time not included in metrics?


> Improve namenode restart times by short-circuiting the first block reports from datanodes
> -----------------------------------------------------------------------------------------
>
>                 Key: HDFS-1295
>                 URL: https://issues.apache.org/jira/browse/HDFS-1295
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: name-node
>    Affects Versions: 0.22.0
>            Reporter: dhruba borthakur
>            Assignee: Matt Foley
>             Fix For: 0.23.0
>
>         Attachments: IBR_shortcut_v2a.patch, IBR_shortcut_v3atrunk.patch, IBR_shortcut_v4atrunk.patch,
IBR_shortcut_v4atrunk.patch, IBR_shortcut_v4atrunk.patch, IBR_shortcut_v6atrunk.patch, shortCircuitBlockReport_1.txt
>
>
> The namenode restart is dominated by the performance of processing block reports. On
a 2000 node cluster with 90 million blocks,  block report processing takes 30 to 40 minutes.
The namenode "diffs" the contents of the incoming block report with the contents of the blocks
map, and then applies these diffs to the blocksMap, but in reality there is no need to compute
the "diff" because this is the first block report from the datanode.
> This code change improves block report processing time by 300%.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message