Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-31445

Avoid floating-point division in millisToDays

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Minor
    • Resolution: Won't Fix
    • 2.4.5
    • None
    • SQL
    • None

    Description

      As the benchmark https://github.com/MaxGekk/spark/pull/27, and comparison to Spark 3.0+an optimisation of fromJavaDate in https://github.com/apache/spark/pull/28205 show that floating-point ops in millisToDays badly impact on the performance of conversion java.sql.Date to Catalyst's date values. The ticket aims to replace double ops by int/long ops.

      Attachments

        Issue Links

          relates to

          Sub-task - The sub-task of the issue SPARK-31439 Perf regression of fromJavaDate

          • Major - Major loss of function.
          • Resolved
          links to

          Web Link GitHub Pull Request #28216

          Activity

            People

              Unassigned Unassigned
              maxgekk Max Gekk
              Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: