You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "haiyangyu (Jira)" <ji...@apache.org> on 2021/01/26 09:11:00 UTC
[jira] [Updated] (SPARK-34242) Use getPartitionByNames to filter
partition to avoid partition scan
[ https://issues.apache.org/jira/browse/SPARK-34242?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
haiyangyu updated SPARK-34242:
------------------------------
Description:
In HiveShim,
{code:java}
getPartitionByFilters{code}
will lead parttion scan in special , ec: concat(dt, phour)='2021-01-26 18', the
{code:java}
convertFilters{code}
can't explain this case to filter partition.
In this issue case , It will
{code:java}
getAllPartitionName{code}
to fetch all partition, and will calc local to filter all quafilied partition, and use getPartitionByNames to fetch partition detail info as last.
> Use getPartitionByNames to filter partition to avoid partition scan
> -------------------------------------------------------------------
>
> Key: SPARK-34242
> URL: https://issues.apache.org/jira/browse/SPARK-34242
> Project: Spark
> Issue Type: New Feature
> Components: SQL
> Affects Versions: 2.4.0, 3.0.0, 3.0.1
> Reporter: haiyangyu
> Priority: Major
>
> In HiveShim,
> {code:java}
> getPartitionByFilters{code}
> will lead parttion scan in special , ec: concat(dt, phour)='2021-01-26 18', the
> {code:java}
> convertFilters{code}
> can't explain this case to filter partition.
> In this issue case , It will
> {code:java}
> getAllPartitionName{code}
> to fetch all partition, and will calc local to filter all quafilied partition, and use getPartitionByNames to fetch partition detail info as last.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org