[IMPALA-5509] Runtime filter : Extend runtime filter to support Dictionary values

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: Impala 2.9.0
• Fix Version/s: Impala 4.1.0
• Component/s: Backend
• Labels: performance runtime-filters
• Epic Color: ghx-label-5
• Assignee: Csaba Ringhofer
• Reporter: Alan Choi
• Created: 14/Jun/17 22:57
• Updated: 18/Nov/24 09:49
• Resolved: 03/Jul/22 12:57

Description

For runtime filter on a single column, it can be run against the dictionary values in Parquet to enable efficient block filtering.

Issue Links

• causes: IMPALA-13193 RuntimeFilter on parquet dictionary should evaluate null values
• is related to: IMPALA-5633 Bloom filters underestimate false positive probability
• is related to: IMPALA-10910 Iceberg scans don't apply runtime filters at Parquet row group level
• is related to: IMPALA-13553 Always evaluate small runtime filters on parquet dictionary
• relates to: IMPALA-6266 Runtime filters should not have non-deterministic expression on consumer side
• relates to: IMPALA-3430 Runtime filter : Extend runtime filter to support Min/Max values for HDFS scans