[ https://issues.apache.org/jira/browse/SPARK-1487?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13999034#comment-13999034 ]
Michael Armbrust commented on SPARK-1487: ----------------------------------------- PR here: https://github.com/apache/spark/pull/511/ > Support record filtering via predicate pushdown in Parquet > ---------------------------------------------------------- > > Key: SPARK-1487 > URL: https://issues.apache.org/jira/browse/SPARK-1487 > Project: Spark > Issue Type: Improvement > Components: SQL > Affects Versions: 1.0.0 > Reporter: Andre Schumacher > Assignee: Andre Schumacher > Fix For: 1.1.0 > > > Parquet has support for column filters, which can be used to avoid reading > and de-serializing records that fail the column filter condition. This can > lead to potentially large savings, depending on the number of columns > filtered by and how many records actually pass the filter. -- This message was sent by Atlassian JIRA (v6.2#6252)