[GitHub] [hadoop-ozone] runzhiwang opened a new pull request #1035: HDDS-3750. Improve SCM performance with 3.2% by avoid stream.collect

Mon, 08 Jun 2020 01:25:20 -0700 


runzhiwang opened a new pull request #1035:
URL: https://github.com/apache/hadoop-ozone/pull/1035


   ## What changes were proposed in this pull request?
    start a ozone cluster with 1000 datanodes and 10 s3gateway, and run two 
weeks with heavy workload, and perf scm. There are 3 places use stream.collect 
which cost 3.2% cpu.
   
   1. stream.collect in PipelineStateMap.getPipelines cost 1.81%
   
![image](https://user-images.githubusercontent.com/51938049/84008272-2f35b100-a9a4-11ea-8104-c12ef283fd19.png)
   
   2. stream.collect in Pipeline.getProtobufMessage cost 0.77%
   
![image](https://user-images.githubusercontent.com/51938049/84008331-4c6a7f80-a9a4-11ea-866d-e47e5c2568c3.png)
   
   3. stream.collect in SCMClientProtocolServer.getContainerWithPipelineBatch 
cost 0.64%
   
![image](https://user-images.githubusercontent.com/51938049/84008396-66a45d80-a9a4-11ea-9cc6-4d4452a7559d.png)
   
   
   ## What is the link to the Apache JIRA
   
   https://issues.apache.org/jira/browse/HDDS-3750
   
   Please replace this section with the link to the Apache JIRA)
   
   Existed tests.
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to