Uploaded image for project: 'Parquet'
  1. Parquet
  2. PARQUET-1841

[C++] Experiment to see if using SIMD shuffle operations for DecodeSpaced improves performance

Attach filesAttach ScreenshotAdd voteVotersWatch issueWatchersCreate sub-taskLinkCloneUpdate Comment AuthorReplace String in CommentUpdate Comment VisibilityDelete Comments
    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Open
    • Major
    • Resolution: Unresolved
    • None
    • None
    • parquet-cpp
    • None

    Description

      Followup from PARQUET-1840 for current benchmarks it seems that doing removing the memset somehow either has no impact or is slightly worse.  We should investigate using SIMD operations to speed up spacing. 

       

      As part of this we can see if moving the memset to only cover uninitialized values after moving all required values provides any speedup.

      Attachments

        Activity

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

          People

            emkornfield@gmail.com Micah Kornfield
            emkornfield@gmail.com Micah Kornfield

            Dates

              Created:
              Updated:

              Slack

                Issue deployment