drill-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Paul Rogers (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (DRILL-5294) Managed External Sort throws an OOM during the merge and spill phase
Date Thu, 23 Feb 2017 01:45:44 GMT

    [ https://issues.apache.org/jira/browse/DRILL-5294?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15879685#comment-15879685
] 

Paul Rogers commented on DRILL-5294:
------------------------------------

Structure of the input batches:

{code}
Actual batch schema & sizes {
  col433(std col. size: 54, actual col. size: 9, total size: 13303, data size: 9207, row capacity:
1023, density: 70)
  EXPR$1(std col. size: 54, actual col. size: 14, total size: 18418, data size: 14322, row
capacity: 1023, density: 78)
  EXPR$2(std col. size: 54, actual col. size: 14, total size: 18418, data size: 14322, row
capacity: 1023, density: 78)
  EXPR$3(std col. size: 54, actual col. size: 3, total size: 7165, data size: 3069, row capacity:
1023, density: 43)
  EXPR$4(std col. size: 54, actual col. size: 6, total size: 10234, data size: 6138, row capacity:
1023, density: 60)
  EXPR$5(std col. size: 54, actual col. size: 10, total size: 14326, data size: 10230, row
capacity: 1023, density: 72)
  EXPR$6(std col. size: 54, actual col. size: 20, total size: 24556, data size: 20460, row
capacity: 1023, density: 84)
  EXPR$7(std col. size: 54, actual col. size: 11, total size: 15349, data size: 11253, row
capacity: 1023, density: 74)
  EXPR$8(std col. size: 54, actual col. size: 13, total size: 17395, data size: 13299, row
capacity: 1023, density: 77)
  EXPR$9(std col. size: 54, actual col. size: 5, total size: 9211, data size: 5115, row capacity:
1023, density: 56)
  EXPR$10(std col. size: 54, actual col. size: 10, total size: 14326, data size: 10230, row
capacity: 1023, density: 72)
  EXPR$11(std col. size: 54, actual col. size: 13, total size: 17395, data size: 13299, row
capacity: 1023, density: 77)
  EXPR$12(std col. size: 54, actual col. size: 14, total size: 18418, data size: 14322, row
capacity: 1023, density: 78)
  EXPR$13(std col. size: 54, actual col. size: 6, total size: 10234, data size: 6138, row
capacity: 1023, density: 60)
  EXPR$14(std col. size: 54, actual col. size: 9, total size: 13303, data size: 9207, row
capacity: 1023, density: 70)
  EXPR$15(std col. size: 54, actual col. size: 10, total size: 14326, data size: 10230, row
capacity: 1023, density: 72)
  EXPR$16(std col. size: 54, actual col. size: 9, total size: 13303, data size: 9207, row
capacity: 1023, density: 70)
  EXPR$17(std col. size: 54, actual col. size: 7, total size: 11257, data size: 7161, row
capacity: 1023, density: 64)
  EXPR$18(std col. size: 54, actual col. size: 20, total size: 24556, data size: 20460, row
capacity: 1023, density: 84)
  EXPR$19(std col. size: 54, actual col. size: 14, total size: 18418, data size: 14322, row
capacity: 1023, density: 78)
  EXPR$20(std col. size: 54, actual col. size: 7, total size: 11257, data size: 7161, row
capacity: 1023, density: 64)
  EXPR$21(std col. size: 54, actual col. size: 4, total size: 8188, data size: 4092, row capacity:
1023, density: 50)
  EXPR$22(std col. size: 54, actual col. size: 9, total size: 13303, data size: 9207, row
capacity: 1023, density: 70)
  EXPR$23(std col. size: 54, actual col. size: 7, total size: 11257, data size: 7161, row
capacity: 1023, density: 64)
  EXPR$24(std col. size: 54, actual col. size: 11, total size: 15349, data size: 11253, row
capacity: 1023, density: 74)
  Records: 1023, Total size: 365313, Row width:359, Density:70}
{code}

All columns are of reasonable size, vectors are small, batches are just 360 K in size. However,
when taking ownership of the batches, memory goes up by 500K as shown earlier. This needs
an explanation.

> Managed External Sort throws an OOM during the merge and spill phase
> --------------------------------------------------------------------
>
>                 Key: DRILL-5294
>                 URL: https://issues.apache.org/jira/browse/DRILL-5294
>             Project: Apache Drill
>          Issue Type: Bug
>          Components: Execution - Relational Operators
>            Reporter: Rahul Challapalli
>            Assignee: Paul Rogers
>             Fix For: 1.10.0
>
>         Attachments: 2751ce6d-67e6-ae08-3b68-e33b29f9d2a3.sys.drill, drillbit.log
>
>
> commit # : 38f816a45924654efd085bf7f1da7d97a4a51e38
> The below query fails with managed sort while it succeeds on the old sort
> {code}
> select * from (select columns[433] col433, columns[0], columns[1],columns[2],columns[3],columns[4],columns[5],columns[6],columns[7],columns[8],columns[9],columns[10],columns[11]
from dfs.`/drill/testdata/resource-manager/3500cols.tbl` order by columns[450],columns[330],columns[230],columns[220],columns[110],columns[90],columns[80],columns[70],columns[40],columns[10],columns[20],columns[30],columns[40],columns[50])
d where d.col433 = 'sjka skjf';
> Error: RESOURCE ERROR: External Sort encountered an error while spilling to disk
> Fragment 1:11
> [Error Id: 0aa20284-cfcc-450f-89b3-645c280f33a4 on qa-node190.qa.lab:31010] (state=,code=0)
> {code}
> Env : 
> {code}
> No of Drillbits : 1
> DRILL_MAX_DIRECT_MEMORY="32G"
> DRILL_MAX_HEAP="4G"
> {code}
> Attached the logs and profile. Data is too large for a jira



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message