[
https://issues.apache.org/jira/browse/HIVE-20079?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16776925#comment-16776925
]
BELUGA BEHR commented on HIVE-20079:
------------------------------------
This patch is still incorrect. It's actually producing the same wrong numbers
as before, though, perhaps a bit more efficiently.
{code}
totalSize += block.getTotalByteSize();
{code}
{{getTotalByteSize()}} is not the same as "rawDataSize".
bq. rawDataSize—Approximate size of data in memory
https://www.cloudera.com/documentation/enterprise/5-15-x/topics/admin_hos_tuning.html
That means that for a single table row with 4 INTs (values: 1,2,3,4) I would
expect a rawDataSize of (4 bytes x 4 Java ints) = 32 bytes. However, Parquet
would report this as 4 bytes because of the way that Parquet packs these
numbers internal to its implementation. Hive should look at the row counts and
multiply it by the row data types.
The {{AbstractSerDe}} class should have code to facilitate all of this like
{{readNumber()}} {{readString(int bumBytes}}, etc that can be called as each
row is read.
> Populate more accurate rawDataSize for parquet format
> -----------------------------------------------------
>
> Key: HIVE-20079
> URL: https://issues.apache.org/jira/browse/HIVE-20079
> Project: Hive
> Issue Type: Improvement
> Components: File Formats
> Affects Versions: 2.0.0
> Reporter: Aihua Xu
> Assignee: Antal Sinkovits
> Priority: Major
> Attachments: HIVE-20079.1.patch, HIVE-20079.2.patch,
> HIVE-20079.3.patch, HIVE-20079.4.patch, HIVE-20079.5.patch, HIVE-20079.6.patch
>
>
> Run the following queries and you will see the raw data for the table is 4
> (that is the number of fields) incorrectly. We need to populate correct data
> size so data can be split properly.
> {noformat}
> SET hive.stats.autogather=true;
> CREATE TABLE parquet_stats (id int,str string) STORED AS PARQUET;
> INSERT INTO parquet_stats values(0, 'this is string 0'), (1, 'string 1');
> DESC FORMATTED parquet_stats;
> {noformat}
> {noformat}
> Table Parameters:
> COLUMN_STATS_ACCURATE true
> numFiles 1
> numRows 2
> rawDataSize 4
> totalSize 373
> transient_lastDdlTime 1530660523
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
