You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@kylin.apache.org by "Temple Zhou (Jira)" <ji...@apache.org> on 2019/09/26 06:54:00 UTC

[jira] [Created] (KYLIN-4180) Prevent abnormal CPU usage by limiting flat filters length

Temple Zhou created KYLIN-4180:
----------------------------------

             Summary: Prevent abnormal CPU usage by limiting flat filters length
                 Key: KYLIN-4180
                 URL: https://issues.apache.org/jira/browse/KYLIN-4180
             Project: Kylin
          Issue Type: Improvement
          Components: Metadata
    Affects Versions: v2.6.0, v2.5.0, v2.4.0
            Reporter: Temple Zhou
            Assignee: Temple Zhou


The following SQL will lead to the list  returned by org.apache.kylin.metadata.filter.TupleFilter#cartesianProduct become very large, and the process to producing the List will use lots of CPU.

{code:sql}
select * from temp.test_filter
where ((a >= 0 and a <= 1)
or    (a >= 1 and a <= 2)
or    (a >= 2 and a <= 3)
...
or    (a >= 300 and a <= 301)
) and ((b >= 0 and b <= 1)
or    (b >= 1 and b <= 2)
or    (b >= 2 and b <= 3)
...
or    (b >= 300 and b <= 301)
)
{code}





--
This message was sent by Atlassian Jira
(v8.3.4#803005)