[
https://issues.apache.org/jira/browse/HDDS-5685?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17407294#comment-17407294
]
Kaijie Chen commented on HDDS-5685:
-----------------------------------
I tested it on a 3-nodes cluster:
A: DataNode, Client
B: DataNode, OM, SCM
C: DataNode
Each node has multiple hard drives.
The network connection between any 2 nodes of them is equal.
The workload is *200* randomly generated files *128MB* each.
I tested 5 times for each configuration here, the time is in *milliseconds*.
||original (default 4M chunk)||streaming (1M chunk)||streaming (4M chunk)||
|44229|42928|53749|
|45342|48035|45119|
|51626|648636|647177|
|39163|55605|52576|
|51459|47861|40078|
Immediately you can notice that the *3rd* streaming test result is abnormal.
I did the tests in the column first order (original 1, 2, 3, 4, 5, streaming1M
1, 2, 3, 4, 5, streaming4M 1, 2, 3, 4, 5),
so it's not related to the workload or network activities at that time.
Despite of this, it seems there is no significant performance improvement here.
Do you have any idea of where possibly it can go wrong?
> Ozone Streaming Performance test.
> ---------------------------------
>
> Key: HDDS-5685
> URL: https://issues.apache.org/jira/browse/HDDS-5685
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: mingchao zhao
> Assignee: mingchao zhao
> Priority: Major
> Attachments: image-2021-08-27-11-07-36-526.png
>
>
> We can test and discuss the performance of Ozone Streaming under this JIRA.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
