[ 
https://issues.apache.org/jira/browse/HDFS-2984?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13439842#comment-13439842
 ] 

Ravi Prakash commented on HDFS-2984:
------------------------------------

Hmm... I ran 

{noformat}
bin/hadoop org.apache.hadoop.fs.slive.SliveTest -duration 9999999 -dirSize 100 
-files 10000 -maps 4 -seed 1 -readSize 20480,20480 -writeSize 20480,20480 
-appendSize 20480,20480 -replication 1,1 -blockSize 10240,10240 -delete 
0,uniform -create 100,uniform -mkdir 0,uniform -rename 0,uniform -append 
0,uniform -ls 0,uniform -read 0,uniform {noformat}
and then 
{noformat} bin/hadoop org.apache.hadoop.fs.slive.SliveTest -duration 9999999 
-dirSize 100 -files 10000 -maps 4 -seed 1 -readSize 20480,20480 -writeSize 
20480,20480 -appendSize 20480,20480 -replication 1,1 -blockSize 10240,10240 
-delete 100,uniform -create 0,uniform -mkdir 0,uniform -rename 0,uniform 
-append 0,uniform -ls 0,uniform -read 0,uniform {noformat}

0.23 seems to be much faster. For 0.20:
Basic report for operation type DeleteOp
-------------
Measurement "failures" = 3212
Measurement "milliseconds_taken" = 11211
Measurement "op_count" = 4000
Measurement "successes" = 788
Rate for measurement "op_count" = 356.792 operations/sec
Rate for measurement "successes" = 70.288 successes/sec
-------------
Basic report for operation type SliveMapper
-------------
Measurement "milliseconds_taken" = 16554
Measurement "op_count" = 4000
Rate for measurement "op_count" = 241.633 operations/sec


whereas for 0.23:
Basic report for operation type DeleteOp
-------------
Measurement "failures" = 3221
Measurement "milliseconds_taken" = 6604
Measurement "op_count" = 4000
Measurement "successes" = 779
Rate for measurement "op_count" = 605.694 operations/sec
Rate for measurement "successes" = 117.959 successes/sec
-------------
Basic report for operation type SliveMapper
-------------
Measurement "milliseconds_taken" = 2456
Measurement "op_count" = 4000
Rate for measurement "op_count" = 1628.664 operations/sec


The command I ran has allowed a much higher %ge of successful operations.
                
> S-live: Rate operation count for delete is worse than 0.20.204 by 28.8%
> -----------------------------------------------------------------------
>
>                 Key: HDFS-2984
>                 URL: https://issues.apache.org/jira/browse/HDFS-2984
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: benchmarks
>    Affects Versions: 0.23.1
>            Reporter: Vinay Kumar Thota
>            Assignee: Ravi Prakash
>            Priority: Critical
>         Attachments: slive.tar.gz
>
>
> Rate operation count for delete is worse than 0.20.204.xx by 28.8%

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to