hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nigel Daley (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-2000) Re-write NNBench to use MapReduce
Date Tue, 16 Oct 2007 16:33:51 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-2000?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12535291
] 

Nigel Daley commented on HADOOP-2000:
-------------------------------------

Looks good.  A few comments:

* import explicit classes, rather than java.io.*
* I'm not sure it makes sense to turn this into a junit test.  As it stands, it won't get
run by the test target because the class doesn't start with the word "Test"
* usage message is missing -bytesToWrite and -operation
* the operation "all" doesn't make sense
* Test Inputs log message missing some parameters
* the barrier() call should go into the mappers map method right before it's ready to start
creating/opening/deleting/renaming and this should be documented so that user set a startTime
that is far enough in the future
* barrier() should throw an exception if the startTime has already passed
* the map should only run 1 operation
* speculation should be turned off explicitly
* "Create a file" comment wrong in doRenameOp and doDeleteOp methods
* doCreateCloseOp should really be doCreateWriteOp and it should create the file with the
appropriate blockSize and replicationFactor and then write the appropriate number of bytes
(which may be zero)
* should there be exception handling around the various file io calls?

> Re-write NNBench to use MapReduce
> ---------------------------------
>
>                 Key: HADOOP-2000
>                 URL: https://issues.apache.org/jira/browse/HADOOP-2000
>             Project: Hadoop
>          Issue Type: Test
>          Components: test
>    Affects Versions: 0.15.0
>            Reporter: Mukund Madhugiri
>            Assignee: Mukund Madhugiri
>             Fix For: 0.16.0
>
>         Attachments: HADOOP-2000.patch, HADOOP-2000.patch
>
>
> The proposal is to re-write the NNBench benchmark/test to measure Namenode operations
using MapReduce. Two buckets of measurements will be done:
> 1. Transactions per second 
> 2. Average latency
> for these operations
> - Create and Close file
> - Open file
> - Rename file
> - Delete file

-- 
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