Kafka
  1. Kafka
  2. KAFKA-741

Improve log cleaning dedupe buffer efficiency

    Details

    • Type: Improvement Improvement
    • Status: Resolved
    • Priority: Major Major
    • Resolution: Duplicate
    • Affects Version/s: None
    • Fix Version/s: 0.8.1
    • Component/s: None
    • Labels:
      None

      Description

      Two good suggestions:
      1. Use a probing scheme to increase density without increasing the collision rate
      2. Only count unique updates to the offset map (i.e. if the key is all zero, don't count it) when computing the load. Dynamically choose the end offset based on when the map is full.

      Would be good to investigate these things.

        Issue Links

          Activity

          Jay Kreps created issue -
          Jay Kreps made changes -
          Field Original Value New Value
          Issue Type Bug [ 1 ] Improvement [ 4 ]
          Assignee Jay Kreps [ jkreps ]
          Fix Version/s 0.8.1 [ 12322960 ]
          Hide
          Jay Kreps added a comment -

          This issue is fixed in the patch for KAFKA-739. It removes the duplication using a probing scheme and counts updates correctly.

          Show
          Jay Kreps added a comment - This issue is fixed in the patch for KAFKA-739 . It removes the duplication using a probing scheme and counts updates correctly.
          Jay Kreps made changes -
          Link This issue duplicates KAFKA-739 [ KAFKA-739 ]
          Jay Kreps made changes -
          Status Open [ 1 ] Resolved [ 5 ]
          Resolution Duplicate [ 3 ]

            People

            • Assignee:
              Jay Kreps
              Reporter:
              Jay Kreps
            • Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development