hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "James Thomas (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-6482) Use block ID-based block layout on datanodes
Date Mon, 09 Jun 2014 21:49:03 GMT

    [ https://issues.apache.org/jira/browse/HDFS-6482?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14025802#comment-14025802
] 

James Thomas commented on HDFS-6482:
------------------------------------

Kihwal, we were considering using some sort of deterministic probing (as in hash tables) to
find less full directories if the initial directory for a block is full. Do you think the
cost (and additional complexity) of this sort of scheme is justified given the relatively
low probability (given the uniform block distribution assumption, at least) of directory blowup?

Additionally, I want to note that if the total number of blocks in the cluster is N, N/2^16
is a strict upper bound on the number of blocks in a single directory on any DN, assuming
completely sequential block IDs. So for a small cluster we can't see any blowup.

> Use block ID-based block layout on datanodes
> --------------------------------------------
>
>                 Key: HDFS-6482
>                 URL: https://issues.apache.org/jira/browse/HDFS-6482
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: datanode
>    Affects Versions: 2.5.0
>            Reporter: James Thomas
>            Assignee: James Thomas
>         Attachments: HDFS-6482.1.patch, HDFS-6482.2.patch, HDFS-6482.patch
>
>
> Right now blocks are placed into directories that are split into many subdirectories
when capacity is reached. Instead we can use a block's ID to determine the path it should
go in. This eliminates the need for the LDir data structure that facilitates the splitting
of directories when they reach capacity as well as fields in ReplicaInfo that keep track of
a replica's location.
> An extension of the work in HDFS-3290.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message