[
https://issues.apache.org/jira/browse/HDFS-2126?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13060293#comment-13060293
]
Matt Foley commented on HDFS-2126:
----------------------------------
The next three sub-tasks parallelize the writing of multiple copies of the
FSImage (HDFS-1071 and HDFS-1826), and greatly improve the efficiency of both
reading and writing the FSImage (HDFS-1070).
Finally, HDFS-1391 proposes a way to speed up the final processing of under-
and over-replicated blocks upon leaving Safe Mode. Implementation of this
improvement is still underway.
These 8 bugs together bring the total startup cost for a 2000-node cluster down
to approximately 20 minutes, assuming adequate network availability and
concurrent startup of all datanodes.
> Improve Namenode startup time [umbrella task]
> ---------------------------------------------
>
> Key: HDFS-2126
> URL: https://issues.apache.org/jira/browse/HDFS-2126
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: name-node
> Affects Versions: 0.20.2
> Reporter: Matt Foley
>
> This is an umbrella task to group the improvements in Namenode startup
> latency made over the last few months, and track remaining ideas.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
