[
https://issues.apache.org/jira/browse/IMPALA-8316?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16796499#comment-16796499
]
ASF subversion and git services commented on IMPALA-8316:
---------------------------------------------------------
Commit 2014199c80c6c74ba4e46139337cd41e1089cc9e in impala's branch
refs/heads/master from Todd Lipcon
[ https://gitbox.apache.org/repos/asf?p=impala.git;h=2014199 ]
IMPALA-8316. Update re2 to the latest version
This updates re2 to the latest tagged release from github.
I benchmarked this with a simple query:
select sum(l_linenumber) from item_20x where
length(regexp_extract(l_shipinstruct, '.*', 0)) > 0
Prior to the change:
- TotalCpuTime: 42s848ms
- wall time: ~19sec
With the change:
- TotalCpuTime: 33s634ms
- wall time: 14-15sec
Change-Id: Id41ca642f5f48fd6237e13f7cab0445e0a402816
Reviewed-on: http://gerrit.cloudera.org:8080/12778
Reviewed-by: Lars Volker <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>
> Update re2 to avoid lock contention
> -----------------------------------
>
> Key: IMPALA-8316
> URL: https://issues.apache.org/jira/browse/IMPALA-8316
> Project: IMPALA
> Issue Type: Improvement
> Components: Backend
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Priority: Major
> Labels: perf
> Fix For: Impala 3.3.0
>
>
> I ran the following test query and found that it spent a lot of time in lock
> contention within the re2 library:
> ```select sum(l_linenumber) from item_20x where
> regexp_extract(l_shipinstruct, '.*E', 0) like '%E' ;```
> I think this lock contention would happen on any regex that involves
> backtracking. This was fixed in the re2 library upstream in
> https://github.com/google/re2/commit/eb00dfdd82015be22086cacc6bf830f72a10e2bc#diff-a60a8d25ed15adf68b94c85775fd3cf7
> We should consider upgrading re2 to the latest release, or if not that, at
> least cherry-picking this perf fix.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
