[
https://issues.apache.org/jira/browse/HBASE-17917?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15994633#comment-15994633
]
Duo Zhang commented on HBASE-17917:
-----------------------------------
Test command
{noformat}
./bin/hbase pe --rows=1000000 --cacheBlocks=false --caching=30
--scanReadType=pread/stream/default --nomapred scan 10
{noformat}
|| ReadType || %us || RS %CPU || DN %CPU ||
| pread | 16 | 400 | 160 |
| stream | 13 | 390 | 30 |
| default | 13 | 390 | 30 |
So for reading the data of the whole region, streaming read will be better(as
now we will open a separated reader for streaming read). The main difference is
at DN side, we can save lots of seeks when using streaming read so the %CPU of
DN will be much less.
Let me run the random scan test.
> Use pread by default for all user scan
> --------------------------------------
>
> Key: HBASE-17917
> URL: https://issues.apache.org/jira/browse/HBASE-17917
> Project: HBase
> Issue Type: Sub-task
> Components: scan
> Affects Versions: 2.0.0
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Fix For: 2.0.0
>
> Attachments: HBASE-17917.patch, HBASE-17917-v1.patch,
> HBASE-17917-v2.patch, HBASE-17917-v2.patch, HBASE-17917-v3.patch,
> HBASE-17917-v4.patch, HBASE-17917-v5.patch
>
>
> As said in the parent issue. We need some benchmark here first.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
