spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Cheng Lian (JIRA)" <j...@apache.org>
Subject [jira] [Resolved] (SPARK-6618) HiveMetastoreCatalog.lookupRelation should use fine-grained lock
Date Tue, 31 Mar 2015 08:51:52 GMT

     [ https://issues.apache.org/jira/browse/SPARK-6618?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Cheng Lian resolved SPARK-6618.
-------------------------------
       Resolution: Fixed
    Fix Version/s: 1.4.0
                   1.3.1

> HiveMetastoreCatalog.lookupRelation should use fine-grained lock
> ----------------------------------------------------------------
>
>                 Key: SPARK-6618
>                 URL: https://issues.apache.org/jira/browse/SPARK-6618
>             Project: Spark
>          Issue Type: Improvement
>          Components: SQL
>    Affects Versions: 1.3.0
>            Reporter: Yin Huai
>            Assignee: Yin Huai
>            Priority: Blocker
>             Fix For: 1.3.1, 1.4.0
>
>
> Right now the entire method of HiveMetastoreCatalog.lookupRelation has a lock (https://github.com/apache/spark/blob/branch-1.3/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveMetastoreCatalog.scala#L173)
and the scope of lock will cover resolving data source tables (https://github.com/apache/spark/blob/branch-1.3/sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveMetastoreCatalog.scala#L93).
So, lookupRelation can be extremely expensive when we are doing expensive operations like
parquet schema discovery. So, we should use fine-grained lock for lookupRelation.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org


Mime
View raw message