[ https://issues.apache.org/jira/browse/BEAM-7929?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16903192#comment-16903192 ]
Rui Wang commented on BEAM-7929: -------------------------------- Calcite has well support on filter/pushdown so we don't need to analyzing Calcite plan and Calcite will help us do that by optimization rules and calling convention. For example, BeamSQL has adopted some rules: https://github.com/apache/beam/blob/master/sdks/java/extensions/sql/src/main/java/org/apache/beam/sdk/extensions/sql/impl/planner/BeamRuleSets.java#L73 I can find many useful links on this topic, like: http://mail-archives.apache.org/mod_mbox/calcite-dev/201610.mbox/%3c96745466-0bfb-49b4-8396-70a992cae...@apache.org%3E https://stackoverflow.com/questions/40217160/how-to-push-down-project-filter-aggregation-to-tablescan-in-calcite There is a also a good slides to cover many piece of calcite: https://www.slideshare.net/JordanHalterman/introduction-to-apache-calcite > ParquetTable.buildIOReader should support column projection and filter > predicate > -------------------------------------------------------------------------------- > > Key: BEAM-7929 > URL: https://issues.apache.org/jira/browse/BEAM-7929 > Project: Beam > Issue Type: New Feature > Components: dsl-sql > Affects Versions: 2.14.0 > Reporter: Neville Li > Priority: Critical > > To leverage the performance improvements in Parquet. -- This message was sent by Atlassian JIRA (v7.6.14#76016)