hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kihwal Lee (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HDFS-5498) Improve datanode startup time
Date Fri, 14 Feb 2014 16:17:22 GMT

     [ https://issues.apache.org/jira/browse/HDFS-5498?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Kihwal Lee updated HDFS-5498:
-----------------------------

    Attachment: HDFS-5498_sh.patch

Attaching a new patch.

[~vinayrpet], you are right. I added a shutdown hook in the new patch, so that the usage values
can be saved on receiving SIGTERM.  We may add a DN-level shutdown hook, but that's beyond
the scope of this jira.

It does not have any functional dependency but the patch was created on top of HDFS-5585,
HDFS-5583 and HDFS-5924. Without them, it will be missing some imports and context may be
wrong.

> Improve datanode startup time
> -----------------------------
>
>                 Key: HDFS-5498
>                 URL: https://issues.apache.org/jira/browse/HDFS-5498
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>            Reporter: Kihwal Lee
>            Assignee: Kihwal Lee
>         Attachments: HDFS-5498.with_du_change.patch, HDFS-5498.with_du_change.patch,
HDFS-5498_sh.patch
>
>
> Similarly to HDFS-5027, an improvement  can be made for getVomeMap(). This is the phase
in which ReplicaMap.is populated.  But it will be even better if datanode scans only once
and do both.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message