hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kai Zheng (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-6133) Make Balancer support exclude specified path
Date Sun, 21 Dec 2014 03:46:13 GMT

    [ https://issues.apache.org/jira/browse/HDFS-6133?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14255032#comment-14255032
] 

Kai Zheng commented on HDFS-6133:
---------------------------------

Hi Yunjiong,

The patch looks great. Minor comments:
1. Regarding the following codes, are you only putting the first DN as favored node or all
the DNs.
{code}
+      InetSocketAddress[] favoredNodes = new InetSocketAddress[numOfDatanodes];
+      for (int i = 0; i < favoredNodes.length; i++) {
+        favoredNodes[i] = cluster.getDataNodes().get(0).getXferAddress();
+      }
{code}

2. In DistributedFileSystem.java, there is a comment as below. It would be better to be updated
as well.
{code}
  /**
   * Same as  
   * {@link #create(Path, FsPermission, boolean, int, short, long, 
   * Progressable)} with the addition of favoredNodes that is a hint to 
   * where the namenode should place the file blocks.
   * The favored nodes hint is not persisted in HDFS. Hence it may be honored
   * at the creation time only. HDFS could move the blocks during balancing or
   * replication, to move the blocks from favored nodes. A value of null means
   * no favored nodes for this create
   */
{code}

3. By the way, a question: do we need any handling for favored nodes and the pin way considering
storage type/policy/Mover?

Thanks.

> Make Balancer support exclude specified path
> --------------------------------------------
>
>                 Key: HDFS-6133
>                 URL: https://issues.apache.org/jira/browse/HDFS-6133
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: balancer & mover, namenode
>            Reporter: zhaoyunjiong
>            Assignee: zhaoyunjiong
>         Attachments: HDFS-6133-1.patch, HDFS-6133-2.patch, HDFS-6133-3.patch, HDFS-6133-4.patch,
HDFS-6133-5.patch, HDFS-6133.patch
>
>
> Currently, run Balancer will destroying Regionserver's data locality.
> If getBlocks could exclude blocks belongs to files which have specific path prefix, like
"/hbase", then we can run Balancer without destroying Regionserver's data locality.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message