[
https://issues.apache.org/jira/browse/HDDS-3703?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Shashikant Banerjee updated HDDS-3703:
--------------------------------------
Fix Version/s: 0.6.0
Resolution: Fixed
Status: Resolved (was: Patch Available)
> Consider avoiding file lookup calls in writeChunk hotpath
> ---------------------------------------------------------
>
> Key: HDDS-3703
> URL: https://issues.apache.org/jira/browse/HDDS-3703
> Project: Hadoop Distributed Data Store
> Issue Type: Improvement
> Reporter: Rajesh Balamohan
> Assignee: Attila Doroszlai
> Priority: Blocker
> Labels: Triaged, performance, pull-request-available
> Fix For: 0.6.0
>
> Attachments: Screenshot 2020-06-02 at 5.28.59 PM.png
>
>
> In getChunkFile internally, it invokes "verifyChunkDirExists". This causes
> file existence checks for the directory and throws IO exception accordingly.
> If the file is anyways going to be written, it is better to handle it later
> and throw the same exception. This could avoid file checks for every
> "writechunk"
> [https://github.com/apache/hadoop-ozone/blob/master/hadoop-hdds/container-service/src/main/java/org/apache/hadoop/ozone/container/keyvalue/impl/FilePerBlockStrategy.java#L106]
>
> File channels are cached anyways in "OpenFiles". So if we can avoid
> "file.getAbsolutePath()", this could save memory and resolving paths.
> [https://github.com/apache/hadoop-ozone/blob/master/hadoop-hdds/container-service/src/main/java/org/apache/hadoop/ozone/container/keyvalue/impl/FilePerBlockStrategy.java#L118]
>
> Also "validateChunkForOverwrite" can be optimised, as "isOverWritePermitted"
> would be false most of the times.
>
> !Screenshot 2020-06-02 at 5.28.59 PM.png|width=835,height=510!
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
