Uploaded image for project: 'Pig'
  1. Pig
  2. PIG-3839 Umbrella jira for Pig on Tez Performance Improvements
  3. PIG-4785

Optimize multi-query plan for diamond shape edges

    XMLWordPrintableJSON

    Details

    • Type: Sub-task
    • Status: Open
    • Priority: Major
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: 0.18.0
    • Component/s: tez
    • Labels:
      None

      Description

      If diamond shaped edge (two edges going to same vertex), we do not merge into Split. Lot of data is transferred because of that. It can be optimized to merge the operator into the Split, but still have a POValueInputTez->POValueOutputTez vertex which just will be used to redirect the input to avoid the diamond shaped edge. This will allow filtering and other processing to happen in the Split operator itself and the data transferred to the routing vertex will be minimal.

        Attachments

          Activity

            People

            • Assignee:
              rohini Rohini Palaniswamy
              Reporter:
              rohini Rohini Palaniswamy
            • Votes:
              1 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated: