Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-6524

Choosing datanode retries times considering with block replica number

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Minor
    • Resolution: Not A Problem
    • 3.0.0-alpha1
    • None
    • hdfs-client

    Description

      Currently the chooseDataNode() does retry with the setting: dfsClientConf.maxBlockAcquireFailures, which by default is 3 (DFS_CLIENT_MAX_BLOCK_ACQUIRE_FAILURES_DEFAULT = 3), it would be better having another option, block replication factor. One cluster with only two block replica setting, or using Reed-solomon encoding solution with one replica factor. It helps to reduce the long tail latency.

      Attachments

        1. HDFS-6524.007.patch
          4 kB
          Lisheng Sun
        2. HDFS-6524.006.patch
          4 kB
          Lisheng Sun
        3. HDFS-6524.005(2).patch
          3 kB
          Lisheng Sun
        4. HDFS-6524.005.patch
          3 kB
          Lisheng Sun
        5. HDFS-6524.004.patch
          4 kB
          Lisheng Sun
        6. HDFS-6524.003.patch
          3 kB
          Lisheng Sun
        7. HDFS-6524.002.patch
          2 kB
          Lisheng Sun
        8. HDFS-6524.001.patch
          2 kB
          Lisheng Sun
        9. HDFS-6524.txt
          2 kB
          Liang Xie

        Activity

          People

            leosun08 Lisheng Sun
            xieliang007 Liang Xie
            Votes:
            0 Vote for this issue
            Watchers:
            7 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: