drill-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kunal Khatua (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (DRILL-6140) Operators listed in Profiles Page doesn't always correspond with operator specified in Physical Plan
Date Wed, 07 Feb 2018 00:13:00 GMT

     [ https://issues.apache.org/jira/browse/DRILL-6140?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Kunal Khatua updated DRILL-6140:
--------------------------------
    Attachment: 25978a1a-24cf-fb4a-17af-59e7115b4fa1.sys.drill

> Operators listed in Profiles Page doesn't always correspond with operator specified in
Physical Plan
> ----------------------------------------------------------------------------------------------------
>
>                 Key: DRILL-6140
>                 URL: https://issues.apache.org/jira/browse/DRILL-6140
>             Project: Apache Drill
>          Issue Type: Bug
>          Components: Web Server
>    Affects Versions: 1.12.0
>            Reporter: Kunal Khatua
>            Assignee: Kunal Khatua
>            Priority: Major
>         Attachments: 25978a1a-24cf-fb4a-17af-59e7115b4fa1.sys.drill
>
>
> A query's physical plan correctly shows
> {code}
>  00-00 Screen : rowType = RecordType(BIGINT EXPR$0): rowcount = 1.0, cumulative cost
= { ...
>    00-01 Project(EXPR$0=[$0]) : rowType = RecordType(BIGINT EXPR$0): rowcount = 1.0,
cumulative cost = { ...
>      00-02 StreamAgg(group=[{}], EXPR$0=[$SUM0($0)]) : rowType = RecordType(BIGINT EXPR$0):
rowcount = 1.0, cumulative cost = { ...
>        00-03 UnionExchange : rowType = RecordType(BIGINT EXPR$0): rowcount = 1.0, cumulative
cost = { ...
>          01-01 StreamAgg(group=[{}], EXPR$0=[COUNT()]) : rowType = RecordType(BIGINT
EXPR$0): rowcount = 1.0, cumulative cost = { ...
>            01-02 Project($f0=[0]) : rowType = RecordType(INTEGER $f0): rowcount = 1.79279253E7,
cumulative cost = ...
>              01-03 Flatten(flattenField=[$1]) : rowType = RecordType(ANY rfsSpecCode,
...
>                01-04 Project(rfsSpecCode=[$1], PUResultsArray=[$2]) : rowType = ...
>                  01-05 SelectionVectorRemover : rowType = RecordType(ANY schemaName,
...
>                    01-06 Filter(condition=[=($0, 'OnyxBlue')]) : rowType = ...
>                      01-07 Project(schemaName=[$0], ITEM=[ITEM($1, 'rfsSpecCode')], ...
>                        01-08 Scan(groupscan=[ParquetGroupScan [entries=[ReadEntryWithPath
[
> {code}
> However, the profile page shows the operators as...
> ||Operator ID || Type || Metrics||
> |00-xx-00 | SCREEN | ... |
> |00-xx-01 | PROJECT | ... |
> |00-xx-02 | STREAMING_AGGREGATE | ... |
> |00-xx-03 | UNORDERED_RECEIVER | ... |
> |01-xx-00 | SINGLE_SENDER | ... |
> |01-xx-01 | STREAMING_AGGREGATE | ... |
> |01-xx-02 | PROJECT | ... |
> |01-xx-03 | SINGLE_SENDER | ... |
> |01-xx-04 | PROJECT | ... |
> |01-xx-05 | SELECTION_VECTOR_REMOVER | ... |
> |01-xx-06 | FILTER | ... |
> |01-xx-07 | PROJECT | ... |
> |01-xx-08 | PARQUET_ROW_GROUP_SCAN | ... |
> As you can see ... {{FLATTEN}}  operator appears as a {{SINGLE_SENDER}} making the profile
hard to interpret.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message