Details
-
Bug
-
Resolution: Fixed
-
Major
-
7.1.0
-
Untriaged
-
1
-
Unknown
-
CX Sprint 259
Description
Currently, the number of parallel scan matches the number of iodevides (i.e., partitions) even for external datasets. When a cluster is configured with a high number of partitions, the parallel scan when reading Parquet files from S3 is stagnated to a smaller number of threads.