[jira] [Issue Comment Deleted] (LOG4J2-1607) Improve performance of SortedArrayStringMap data structure

["Remko Popma (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/LOG4J2-1607?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Remko Popma updated LOG4J2-1607:
--------------------------------
    Comment: was deleted

(was: When I ran the JMH ThreadContextBenchmark I found a performance problem when merging context data with properties. This is now addressed. Updated benchmark results follow below.

!InjectWithoutProperties.png!
5: context data has 5 elements
50: context data has 50 elements
500: context data has 500 elements

!InjectWithProperties.png!
Same as above, with 5 properties that need to be merged with the context data.

!ThreadContext-PutAndRemove.png!

!ThreadContext-Get.png!)

> Improve performance of SortedArrayStringMap data structure
> ----------------------------------------------------------
>
>                 Key: LOG4J2-1607
>                 URL: https://issues.apache.org/jira/browse/LOG4J2-1607
>             Project: Log4j 2
>          Issue Type: Improvement
>    Affects Versions: 2.7
>            Reporter: Remko Popma
>            Assignee: Remko Popma
>
> [Performance testing|https://issues.apache.org/jira/browse/LOG4J2-1349?focusedCommentId=15432877&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15432877] for LOG4J2-1349 showed that the SortedArray-based data structure can be copied fast, but its {{get()}} and {{put()}} performance is worse than a hash-based data structure like the JDK HashMap.
> This ticket is to investigate options for improving performance of the {{get()}} operation, since this can impact the performance of Filters and Lookups.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: log4j-dev-unsubscribe@logging.apache.org
For additional commands, e-mail: log4j-dev-help@logging.apache.org