[IMPALA-3430] Runtime filter : Extend runtime filter to support Min/Max values for HDFS scans

• Type: New Feature
• Status: Resolved
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: Impala 2.6.0
• Fix Version/s: Impala 4.0.1
• Component/s: Backend
• Labels: performance runtime-filters
• Target Version: Product Backlog
• Assignee: Qifan Chen
• Reporter: Mostafa Mokhtar
• Created: 26/Apr/16 23:04
• Updated: 23/Aug/21 14:16
• Resolved: 23/Aug/21 14:16

Description

Annotating Runtime filters with Min/Max values can help with

• Inequality joins
• Pushing more efficient filters to the scan
• Used to skip reading Parquet blocks reducing IO.

Issue Links

• is related to: IMPALA-5509 Runtime filter : Extend runtime filter to support Dictionary values
• is related to: IMPALA-10325 Parquet scan should use min/max statistics to skip pages based on equi-join predicate
• is related to: IMPALA-5843 Use page index in Parquet files to skip pages
• is related to: IMPALA-4252 Add RuntimeFilters for min/max at KuduScanNode
• is related to: IMPALA-10869 Extend min/max filter for inequality subquery for other use cases
• is related to: IMPALA-4989 Improve filtering based on parquet::Statistics
• is related to: IMPALA-5789 Prune all partitions if an always-false runtime filter is received