[ 
https://issues.apache.org/jira/browse/HBASE-22387?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16850290#comment-16850290
 ] 

Zheng Hu edited comment on HBASE-22387 at 5/29/19 1:48 AM:
-----------------------------------------------------------

After enable the short-circuit read feature, the throughtput improved much.  
see the graphs 
1. 
[G1GC-stage-stw.case02-with-buffer-size-64KB.png|https://issues.apache.org/jira/secure/attachment/12970103/G1GC-stage-stw.case02-with-buffer-size-64KB.png]
 ; 
2. 
[G1GC-stw.case02-with-buffer-size-64KB.png|https://issues.apache.org/jira/secure/attachment/12970101/G1GC-stw.case02-with-buffer-size-64KB.png]
3. 
[QPS-latency.case02-with-buffer-size-64KB.png|https://issues.apache.org/jira/secure/attachment/12970100/QPS-latency.case02-with-buffer-size-64KB.png]
We use the default hbase.ipc.server.allocator.buffer.size=64KB here,  as 
HBASE-22483 said, I will have a benchmark by setting the buffer.size=65KB, 
should be better performance. 


was (Author: openinx):
After enable the short-circuit read feature, the throughtput improved much.  
see the graphs 
1. 
[G1GC-stage-stw.case02-with-buffer-size-64KB.pnghttps://issues.apache.org/jira/secure/attachment/12970103/G1GC-stage-stw.case02-with-buffer-size-64KB.png]
 ; 
2. 
[G1GC-stw.case02-with-buffer-size-64KB.png|https://issues.apache.org/jira/secure/attachment/12970101/G1GC-stw.case02-with-buffer-size-64KB.png]
3. 
[QPS-latency.case02-with-buffer-size-64KB.png|https://issues.apache.org/jira/secure/attachment/12970100/QPS-latency.case02-with-buffer-size-64KB.png]
We use the default hbase.ipc.server.allocator.buffer.size=64KB here,  as 
HBASE-22483 said, I will have a benchmark by setting the buffer.size=65KB, 
should be better performance. 

> Evaluate the get/scan performance after reading HFile block into offheap 
> directly
> ---------------------------------------------------------------------------------
>
>                 Key: HBASE-22387
>                 URL: https://issues.apache.org/jira/browse/HBASE-22387
>             Project: HBase
>          Issue Type: Sub-task
>            Reporter: Zheng Hu
>            Assignee: Zheng Hu
>            Priority: Major
>         Attachments: G1GC-stage-stw.case01.png, 
> G1GC-stage-stw.case02-with-buffer-size-64KB.png, G1GC-stw.case01.png, 
> G1GC-stw.case02-with-buffer-size-64KB.png, QPS-latency.case01.png, 
> QPS-latency.case02-with-buffer-size-64KB.png, blockReaderRemote.png, 
> blocksByHFile-stack-trace.png, screenshot-1.png, 
> test-cluster-configuration-details.png
>
>
> Now, all sub-tasks has been resolved now (except the HBASE-21946 because of 
> the hadoop dependency problem), will provide some performance benckmarks to 
> show the latency improvement.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to