You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@kylin.apache.org by "hujiahua (Jira)" <ji...@apache.org> on 2022/01/10 09:33:00 UTC
[jira] [Updated] (KYLIN-5149) isExactlyMatchSet should keep same size with cuboidIds
[ https://issues.apache.org/jira/browse/KYLIN-5149?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
hujiahua updated KYLIN-5149:
----------------------------
Description:
In some query case, I find the size of `isExactlyMatchSet` was bigger than `icuboidIds`i in query log.
For example, there is a join scene like this :
==========================[QUERY]===============================
Query Id: d99bdb35-5d97-b0c0-35a9-db933f3cb64f
SQL: select a.{*},b.{*} from
(select SELLER_ID,sum(PRICE) as PRICE from KYLIN_SALES where PART_DT > '2012-01-01' and PART_DT < '2012-02-01' GROUP BY SELLER_ID limit 10) a
left join KYLIN_ACCOUNT b on a.SELLER_ID = b.ACCOUNT_ID
User: ADMIN
Success: true
Duration: 19.011
Project: default
Realization Names: [CUBE[name=my_test2_3], CUBE[name=account_info_t1]]
Cuboid Ids: [6, 31]
Is Exactly Matched: [false, false, false]
Total scan count: 10454
Total scan files: 2
Total metadata time: 4ms
Total spark scan time: 0ms
Total scan bytes: 55822
Total return rows: 0
Result row count: 10
Storage cache used: false
Is Query Push-Down: false
Is Prepare: false
Used Spark pool: lightweight_tasks
Trace URL: null
Message: null
Remote Address: 127.0.0.1
was:
In some query case, I find the size of `isExactlyMatchSet`i will bigger than `icuboidIds`i in query log.
For example, there is a join scene like this :
==========================[QUERY]===============================
Query Id: d99bdb35-5d97-b0c0-35a9-db933f3cb64f
SQL: select a.*,b.* from
(select SELLER_ID,sum(PRICE) as PRICE from KYLIN_SALES where PART_DT > '2012-01-01' and PART_DT < '2012-02-01' GROUP BY SELLER_ID limit 10) a
left join KYLIN_ACCOUNT b on a.SELLER_ID = b.ACCOUNT_ID
User: ADMIN
Success: true
Duration: 19.011
Project: default
Realization Names: [CUBE[name=my_test2_3], CUBE[name=account_info_t1]]
Cuboid Ids: [6, 31]
Is Exactly Matched: [false, false, false]
Total scan count: 10454
Total scan files: 2
Total metadata time: 4ms
Total spark scan time: 0ms
Total scan bytes: 55822
Total return rows: 0
Result row count: 10
Storage cache used: false
Is Query Push-Down: false
Is Prepare: false
Used Spark pool: lightweight_tasks
Trace URL: null
Message: null
Remote Address: 127.0.0.1
> isExactlyMatchSet should keep same size with cuboidIds
> ------------------------------------------------------
>
> Key: KYLIN-5149
> URL: https://issues.apache.org/jira/browse/KYLIN-5149
> Project: Kylin
> Issue Type: Bug
> Components: Query Engine
> Affects Versions: v4.0.1
> Reporter: hujiahua
> Priority: Minor
>
> In some query case, I find the size of `isExactlyMatchSet` was bigger than `icuboidIds`i in query log.
>
> For example, there is a join scene like this :
> ==========================[QUERY]===============================
> Query Id: d99bdb35-5d97-b0c0-35a9-db933f3cb64f
> SQL: select a.{*},b.{*} from
> (select SELLER_ID,sum(PRICE) as PRICE from KYLIN_SALES where PART_DT > '2012-01-01' and PART_DT < '2012-02-01' GROUP BY SELLER_ID limit 10) a
> left join KYLIN_ACCOUNT b on a.SELLER_ID = b.ACCOUNT_ID
> User: ADMIN
> Success: true
> Duration: 19.011
> Project: default
> Realization Names: [CUBE[name=my_test2_3], CUBE[name=account_info_t1]]
> Cuboid Ids: [6, 31]
> Is Exactly Matched: [false, false, false]
> Total scan count: 10454
> Total scan files: 2
> Total metadata time: 4ms
> Total spark scan time: 0ms
> Total scan bytes: 55822
> Total return rows: 0
> Result row count: 10
> Storage cache used: false
> Is Query Push-Down: false
> Is Prepare: false
> Used Spark pool: lightweight_tasks
> Trace URL: null
> Message: null
> Remote Address: 127.0.0.1
--
This message was sent by Atlassian Jira
(v8.20.1#820001)