[
https://issues.apache.org/jira/browse/HDFS-14357?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16883688#comment-16883688
]
Rakesh R commented on HDFS-14357:
---------------------------------
Thanks [~PhiloHe] for the patch.
Please take care below comment:
bq. One depends on PMDK libs and the other doesn't. PMDK can bring user
performance gain for cache write and cache read
Please rephrase it like,
The default is based on pure Java implementation and the other is native
implementation which leverages PMDK library to improve the performance of cache
write and cache read.
To enable and use PMDK, use following steps:
1. Build PMDK library. Please refer to the official site "" for detail
information.
2. Build Hadoop with PMDK support. Please refer to `BUILDING.txt` to build
hadoop with PMDK support.
3. Copy PMDK library. Make sure PMDK is available on HDFS DataNodes.
To verify that PMDK is correctly detected by Hadoop, run the hadoop
{{checknative}} command.
bq. For multiply volumes
Typo: {{For multiply volumes}} --> to --> {{For multiple volumes}}
> Update the relevant docs for HDFS cache on SCM support
> ------------------------------------------------------
>
> Key: HDFS-14357
> URL: https://issues.apache.org/jira/browse/HDFS-14357
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Feilong He
> Assignee: Feilong He
> Priority: Major
> Fix For: 3.3.0
>
> Attachments: HDFS-14357.000.patch, HDFS-14357.001.patch,
> HDFS-14357.002.patch, HDFS-14357.003.patch
>
>
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
