Details

    • Type: Improvement Improvement
    • Status: Open
    • Priority: Major Major
    • Resolution: Unresolved
    • Affects Version/s: None
    • Fix Version/s: None
    • Component/s: datanode, performance
    • Labels:
      None

      Description

      In benchmarking HDFS-941 I noticed that for the random read workload, the FSDataset lock is highly contended. After converting it to a ReentrantReadWriteLock, I saw a ~25% improvement on both latency and ops/second.

      1. hdfs-1148-old.txt
        12 kB
        Todd Lipcon
      2. hdfs-1148-trunk.txt
        46 kB
        Todd Lipcon
      3. patch-HDFS-1148-rel0.20.2.txt
        12 kB
        Dave Thompson

        Issue Links

          Activity

          No work has yet been logged on this issue.

            People

            • Assignee:
              Dave Thompson
              Reporter:
              Todd Lipcon
            • Votes:
              0 Vote for this issue
              Watchers:
              22 Start watching this issue

              Dates

              • Created:
                Updated:

                Development