[
https://issues.apache.org/jira/browse/DRILL-3770?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14968097#comment-14968097
]
Deneche A. Hakim commented on DRILL-3770:
-----------------------------------------
Fixing DRILL-3952 should resolve this issue as the window function operator
will no longer require all batches for a particular partition to be held in
memory.
Please note that for some specific window functions (e.g. percent_rank) there
is no other alternative than to wait for all batches for a partition, thus the
fix won't work for them. For such queries, we'll most likely need a spilling
mechanism
> Query with window function having just ORDER BY clause runs out of memory on
> large datasets
> -------------------------------------------------------------------------------------------
>
> Key: DRILL-3770
> URL: https://issues.apache.org/jira/browse/DRILL-3770
> Project: Apache Drill
> Issue Type: Bug
> Components: Execution - Flow
> Affects Versions: 1.2.0
> Environment: 10 Performance Nodes
> DRILL_MAX_DIRECT_MEMORY=100g
> DRILL_INIT_HEAP="8g"
> DRILL_MAX_HEAP="8g"
> planner.memory.query_max_memory_per_node bumped up to 20 GB
> TPC-DS SF 1000 dataset (Parquet)
> Reporter: Abhishek Girish
> Priority: Critical
> Labels: window_function
> Fix For: Future
>
> Attachments: drillbit.log.txt, profile.json
>
>
> The following query runs out of memory:
> {code:sql}
> SELECT SUM(ss.ss_net_paid_inc_tax) OVER (ORDER BY ss.ss_store_sk) FROM
> store_sales ss LIMIT 20;
> java.lang.RuntimeException: java.sql.SQLException: RESOURCE ERROR: One or
> more nodes ran out of memory while executing the query.
> Fragment 0:0
> [Error Id: 9c441211-65ec-4206-9e6b-d6ae9c2903be on ucs-node6.perf.lab:31010]
> at sqlline.IncrementalRows.hasNext(IncrementalRows.java:73)
> at
> sqlline.TableOutputFormat$ResizingRowsProvider.next(TableOutputFormat.java:87)
> at sqlline.TableOutputFormat.print(TableOutputFormat.java:118)
> at sqlline.SqlLine.print(SqlLine.java:1583)
> at sqlline.Commands.execute(Commands.java:852)
> at sqlline.Commands.sql(Commands.java:751)
> at sqlline.SqlLine.dispatch(SqlLine.java:738)
> at sqlline.SqlLine.begin(SqlLine.java:612)
> at sqlline.SqlLine.start(SqlLine.java:366)
> at sqlline.SqlLine.main(SqlLine.java:259)
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
