Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-39634

Allow file splitting in combination with row index generation

Log workAgile BoardRank to TopRank to BottomAttach filesAttach ScreenshotBulk Copy AttachmentsBulk Move AttachmentsAdd voteVotersWatch issueWatchersCreate sub-taskConvert to sub-taskLinkCloneLabelsUpdate Comment AuthorReplace String in CommentUpdate Comment VisibilityDelete Comments
    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Open
    • Major
    • Resolution: Unresolved
    • 3.3.0
    • None
    • SQL
    • None

    Description

      This issue is a follow up for SPARK-37980

      Because of a bug in parquet-mr https://issues.apache.org/jira/browse/PARQUET-2161 it is currently impossible to generate row indexes for parquet files if they are split into multiple pieces. Instead, each file must be read in a single task.

      Once the version of parquet-mr with the fix is included in Spark, we should remove the workarounds from the code (marked with this ticket number) from the code, so that parquet files are splittable even when the row indexes need to be generated.

      Attachments

        Issue Links

        Activity

          This comment will be Viewable by All Users Viewable by All Users
          Cancel

          People

            Unassigned Unassigned Assign to me
            ala.luszczak Ala Luszczak

            Dates

              Created:
              Updated:

              Slack

                Issue deployment