tanvipenumudy opened a new pull request, #5274:
URL: https://github.com/apache/ozone/pull/5274

   ## What changes were proposed in this pull request?
   
   An analysis as described 
[here](https://issues.apache.org/jira/browse/HDDS-9228) has revealed a 
significant S3G read performance issue. This issue became apparent when 
comparing read operations to writes, as the reads exhibited notably slower 
performance and resulted in higher CPU usage. 
   
   JFR:
   ```
   Stack Trace  Count   Percentage
   void org.eclipse.jetty.server.HttpOutput.write(int)  431146  39 %
   void 
org.glassfish.jersey.servlet.internal.ResponseWriter$NonCloseableOutputStreamWrapper.write(int)
 431145  39 %
   void org.glassfish.jersey.message.internal.CommittingOutputStream.write(int) 
431145  39 %
   void java.io.FilterOutputStream.write(int)   431145  39 %
   void java.io.FilterOutputStream.write(byte[], int, int)      431145  39 %
   void 
org.glassfish.jersey.message.internal.WriterInterceptorExecutor$UnCloseableOutputStream.write(byte[],
 int, int) 431145  39 %
   long org.apache.commons.io.IOUtils.copyLarge(InputStream, OutputStream, 
byte[])      431145  39 %
   ```
   
   The root of the problem was traced to how S3G writes its responses, 
`FilterOutputStream.write(byte[], int, int)` transitions to 
`FilterOutputStream.write(int)` handling incoming data arrays as individual 
bytes rather than as complete arrays. The proposed solution involves removing 
the `FilterOutputStream` responsible for this behaviour which would result in 
5x better throughput and CPU usage around 12%.
   
   ## What is the link to the Apache JIRA
   
   https://issues.apache.org/jira/browse/HDDS-9228
   
   ## How was this patch tested?
   
   N/A (trivial change)
   


-- 
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.

To unsubscribe, e-mail: [email protected]

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