[
https://issues.apache.org/jira/browse/HDDS-14245?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Swaminathan Balachandran updated HDDS-14245:
--------------------------------------------
Description: Currently in prepare batch to verify if an operation is still
valid an O(n)check is performed on each and every deleteRange op that has
occured after the operation. This is done for each operation in the batch and
if all deleteRanges are going to be stacked at the end then the prepare batch
can become an O(n{^}2{^}) operation making it unoptimal. This patch aims to
create Index data structure to track all the deleteRange operations and provide
an efficient way to check if a delete range operation exists or not. The delete
ranges would also be popped out of the index structure as and when the delete
range op gets executed. (was: Currently in prepare batch to verify if an
operation is still valid an O(n) check is performed on each and every
deleteRange op that has occured after the operation. This is done for each
operation in the batch and if all deleteRanges are going to be stacked at the
end then the prepare batch can become an O(n{^}2{^}) operation making it
unoptimal. This patch aims to create Index data structure to track all the
deleteRange operations and provide an efficient way to check if a delete range
operation exists or not. The delete ranges would also be popped out of the
index structure as and when the delete range op gets executed.)
> Optimize Prepare Batch looping when there a lot of DeleteRange operations in
> the batch
> --------------------------------------------------------------------------------------
>
> Key: HDDS-14245
> URL: https://issues.apache.org/jira/browse/HDDS-14245
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: Swaminathan Balachandran
> Assignee: Swaminathan Balachandran
> Priority: Major
>
> Currently in prepare batch to verify if an operation is still valid an
> O(n)check is performed on each and every deleteRange op that has occured
> after the operation. This is done for each operation in the batch and if all
> deleteRanges are going to be stacked at the end then the prepare batch can
> become an O(n{^}2{^}) operation making it unoptimal. This patch aims to
> create Index data structure to track all the deleteRange operations and
> provide an efficient way to check if a delete range operation exists or not.
> The delete ranges would also be popped out of the index structure as and when
> the delete range op gets executed.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
