[jira] [Commented] (AURORA-1283) Implement a benchmark for status update processing throughput.

["Benjamin Mahler (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/AURORA-1283?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14516085#comment-14516085 ] 

Benjamin Mahler commented on AURORA-1283:
-----------------------------------------

https://reviews.apache.org/r/33608/

> Implement a benchmark for status update processing throughput.
> --------------------------------------------------------------
>
>                 Key: AURORA-1283
>                 URL: https://issues.apache.org/jira/browse/AURORA-1283
>             Project: Aurora
>          Issue Type: Task
>          Components: Stress Testing
>            Reporter: Benjamin Mahler
>            Assignee: Benjamin Mahler
>
> In order to guide changes for improving status update throughput, a throughput benchmark can provide the data needed to justify structural changes to the code.
> JMH seems to easily support benchmarks in which "synchronous" methods are called concurrently from multiple threads (which works for the current synchronous update code), but it is a bit trickier to leverage JMH for "asynchronous" code (if we were to process updates asynchronously, we become less interested in how long the {{statusUpdate(...)}} call takes, and more interested in the overall time it takes to process large numbers of updates).



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