hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "lohit vijayarenu (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-2559) DFS should place one replica per rack
Date Wed, 05 Mar 2008 10:57:40 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-2559?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12575296#action_12575296
] 

lohit vijayarenu commented on HADOOP-2559:
------------------------------------------

Here are the test results 

1. This test was 200 node random writer followed by just Scan.
{noformat}
Job             Trunk     Patch1    Patch2
RandWriter      460       471       520
Scan            718       583       601
MapTasks        16307     16307     16241
Data-Local Maps 15743     15741     15741
Rack-Local Maps 86        84        54
{noformat}

2. This test was 10 mappers on 200 node writing random text (total 2TB) followed by 200 node
Scan
{noformat}
Job             Trunk     Patch1    Patch2
RandWriter      13633     12923     12299
Scan            764       606       600
MapTasks        16517     16453     16538
Data-Local Maps 15276     15335     15695
Rack-Local Maps 433       363       121
{noformat}


> DFS should place one replica per rack
> -------------------------------------
>
>                 Key: HADOOP-2559
>                 URL: https://issues.apache.org/jira/browse/HADOOP-2559
>             Project: Hadoop Core
>          Issue Type: Improvement
>          Components: dfs
>            Reporter: Runping Qi
>            Assignee: lohit vijayarenu
>         Attachments: HADOOP-2559-1.patch, HADOOP-2559-2.patch, Trunk_Block_Report.png
>
>
> Currently, when writing out a block, dfs will place one copy to a local data node, one
copy to a rack local node
> and another one to a remote node. This leads to a number of undesired properties:
> 1. The block will be rack-local to two tacks instead of three, reducing the advantage
of rack locality based scheduling by 1/3.
> 2. The Blocks of a file (especiallya  large file) are unevenly distributed over the nodes:
One third will be on the local node, and two thirds on the nodes on the same rack. This may
make some nodes full much faster than others, 
> increasing the need of rebalancing. Furthermore, this also make some nodes become "hot
spots" if those big 
> files are popular and accessed by many applications.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message