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-5266) Parquet Reader produces "low density" record batches
Date Wed, 15 Feb 2017 17:41:41 GMT

    [ https://issues.apache.org/jira/browse/DRILL-5266?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15868233#comment-15868233
] 

Paul Rogers commented on DRILL-5266:
------------------------------------

To follow up on the low-density batch issue, Parquet tools reports that the file has a single
row group:

{code}
row group 1:              RC: 1,434,519 TS: 481,347,650 OFFSET:4 
{code}

The two numbers are row count and total (uncompressed) file size.

So, the low-density batch issue is not due to multiple row groups in the file as there is
just one group. The file itself is of this size:

{code}
165,002,354 1_0_0.parquet
{code}

If each batch is just 5% full, we will need 481 MB * 20 = 9.6 GB to buffer the data for sort.
Obviously, this will impact performance negatively.

> Parquet Reader produces "low density" record batches
> ----------------------------------------------------
>
>                 Key: DRILL-5266
>                 URL: https://issues.apache.org/jira/browse/DRILL-5266
>             Project: Apache Drill
>          Issue Type: Bug
>          Components: Storage - Parquet
>    Affects Versions: 1.10
>            Reporter: Paul Rogers
>
> Testing with the managed sort revealed that, for at least one file, Parquet produces
"low-density" batches: batches in which only 5% of each value vector contains actual data,
with the rest being unused space. When fed into the sort, we end up buffering 95% of wasted
space, using only 5% of available memory to hold actual query data. The result is poor performance
of the sort as it must spill far more frequently than expected.
> The managed sort analyzes incoming batches to prepare good memory use estimates. The
following the the output from the Parquet file in question:
> {code}
> Actual batch schema & sizes {
>   T1¦¦cs_sold_date_sk(std col. size: 4, actual col. size: 4, total size: 196608, vector
size: 131072, data size: 4516, row capacity: 32768, density: 4)
>   T1¦¦cs_sold_time_sk(std col. size: 4, actual col. size: 4, total size: 196608, vector
size: 131072, data size: 4516, row capacity: 32768, density: 4)
>   T1¦¦cs_ship_date_sk(std col. size: 4, actual col. size: 4, total size: 196608, vector
size: 131072, data size: 4516, row capacity: 32768, density: 4)
> ...
>   c_email_address(std col. size: 54, actual col. size: 27, total size: 53248, vector
size: 49152, data size: 30327, row capacity: 4095, density: 62)
>   Records: 1129, Total size: 32006144, Row width:28350, Density:5}
> {code}



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

Mime
View raw message