ZhangYao has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/13721 )

Change subject: KUDU-2847: Optimize iteration over selection vector in 
SerializeRowBlock
......................................................................


Patch Set 1:

Maybe my description is a little bit confusing.
The total row count in benchmark is 100000.
The horizontal axis "schema size" is the column count in schema, here
we have 3 columns, 30 columns and 300 columns for test.
The vertical axis "Select rate" decide how many rows we randomly selected,
here we have 1.0(all rows) * 100000, 0.8 * 100000, 0.5(half rows) * 100000
and 0.2 * 100000 for test.


--
To view, visit http://gerrit.cloudera.org:8080/13721
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I19917d1875c46fd4cf98ef8a471b0340a76161e7
Gerrit-Change-Number: 13721
Gerrit-PatchSet: 1
Gerrit-Owner: ZhangYao <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-Reviewer: Yao Xu <[email protected]>
Gerrit-Reviewer: ZhangYao <[email protected]>
Gerrit-Comment-Date: Thu, 27 Jun 2019 12:38:12 +0000
Gerrit-HasComments: No

Reply via email to