Details

    • Type: Sub-task Sub-task
    • Status: Closed
    • Priority: Major Major
    • Resolution: Won't Fix
    • Affects Version/s: None
    • Fix Version/s: None
    • Component/s: None
    • Labels:
      None

      Description

      Use queuing to improve efficiency of block reports.

        Activity

        Transition Time In Source Status Execution Times Last Executer Last Execution Date
        Open Open Resolved Resolved
        68d 23h 12m 1 Sanjay Radia 03/Jan/08 21:52
        Resolved Resolved Closed Closed
        20h 41m 1 Doug Cutting 04/Jan/08 18:33
        Owen O'Malley made changes -
        Project Hadoop Common [ 12310240 ] HDFS [ 12310942 ]
        Key HADOOP-2111 HDFS-393
        Affects Version/s 0.14.0 [ 12312474 ]
        Component/s dfs [ 12310710 ]
        Doug Cutting made changes -
        Status Resolved [ 5 ] Closed [ 6 ]
        Nigel Daley made changes -
        Fix Version/s 0.16.0 [ 12312740 ]
        Sanjay Radia made changes -
        Status Open [ 1 ] Resolved [ 5 ]
        Resolution Won't Fix [ 2 ]
        Hide
        Sanjay Radia added a comment -

        See my last comment

        Show
        Sanjay Radia added a comment - See my last comment
        Sanjay Radia made changes -
        Issue Type Improvement [ 4 ] Sub-task [ 7 ]
        Parent HADOOP-2448 [ 12384810 ]
        Hide
        Sanjay Radia added a comment -

        I implemented this an did some performance benchmarks.
        It did help initial block report processing and eliminated duplicated block-reports due to timeout.
        However, when I combined this with the random backoff (hadoop-2326) it did not help.

        Furthermore, once we cleaniup the RPC layer to remove the client-side timeouts and retries, this work will be obsolete and there
        does not seem to be a good reason to have a dedicated queue and thread for the block report rpc.

        Show
        Sanjay Radia added a comment - I implemented this an did some performance benchmarks. It did help initial block report processing and eliminated duplicated block-reports due to timeout. However, when I combined this with the random backoff (hadoop-2326) it did not help. Furthermore, once we cleaniup the RPC layer to remove the client-side timeouts and retries, this work will be obsolete and there does not seem to be a good reason to have a dedicated queue and thread for the block report rpc.
        Robert Chansler made changes -
        Field Original Value New Value
        Assignee Robert Chansler [ chansler ] Sanjay Radia [ sanjay.radia ]
        Robert Chansler created issue -

          People

          • Assignee:
            Sanjay Radia
            Reporter:
            Robert Chansler
          • Votes:
            0 Vote for this issue
            Watchers:
            0 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development