[
https://issues.apache.org/jira/browse/HDFS-17574?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17864169#comment-17864169
]
ASF GitHub Bot commented on HDFS-17574:
---------------------------------------
granewang opened a new pull request, #6931:
URL: https://github.com/apache/hadoop/pull/6931
Description of PR
As of now,We can not specify data units while specifying the -blockSize
as arguments(like 1m), but have to specify numbers.
while which is support with configuration item dfs.blocksize
How was this patch tested?
Test with command:hadoop
org.apache.hadoop.hdfs.server.namenode.NNThroughputBenchmark -fs
[hdfs://](hdfs://ctyunns/)xxxxx -op create -threads 100 -files 250000
-filesPerDir 100 -blockSize 1m -close
> Make NNThroughputBenchmark support argument blockSize suffix with k, m, g,
> t, p, e
> -----------------------------------------------------------------------------------
>
> Key: HDFS-17574
> URL: https://issues.apache.org/jira/browse/HDFS-17574
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: benchmarks, hdfs
> Affects Versions: 3.3.3, 3.3.6
> Reporter: wangzhongwei
> Assignee: wangzhongwei
> Priority: Major
>
> As of now,We can not specify data units while specifying the -blockSize as
> arguments(like 1m), but have to specify numbers
> test command:
> hadoop org.apache.hadoop.hdfs.server.namenode.NNThroughputBenchmark -fs
> [hdfs://|hdfs://ctyunns/]xxxxx -op create -threads 100 -files 250000
> -filesPerDir 100 -blockSize 1m -close
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
