Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-5238 Improve NN operation throughput
  3. HDFS-7967

Reduce the performance impact of the balancer

    XMLWordPrintableJSON

Details

    • Sub-task
    • Status: Patch Available
    • Critical
    • Resolution: Unresolved
    • 2.0.0-alpha
    • None
    • namenode
    • None

    Description

      The balancer needs to query for blocks to move from overly full DNs. The block lookup is extremely inefficient. An iterator of the node's blocks is created from the iterators of its storages' blocks. A random number is chosen corresponding to how many blocks will be skipped via the iterator. Each skip requires costly scanning of triplets.

      The current design also only considers node imbalances while ignoring imbalances within the nodes's storages. A more efficient and intelligent design may eliminate the costly skipping of blocks via round-robin selection of blocks from the storages based on remaining capacity.

      Attachments

        1. HDFS-7967.branch-2.8.003.patch
          36 kB
          Daryn Sharp
        2. HDFS-7967.branch-2.002.patch
          33 kB
          Daryn Sharp
        3. HDFS-7967.branch-2.8.002.patch
          36 kB
          Daryn Sharp
        4. HDFS-7967.branch-2.001.patch
          33 kB
          Daryn Sharp
        5. HDFS-7967.branch-2.8.001.patch
          35 kB
          Daryn Sharp
        6. HDFS-7967.branch-2-1.patch
          33 kB
          Daryn Sharp
        7. HDFS-7967.branch-2.8-1.patch
          35 kB
          Daryn Sharp
        8. HDFS-7967-branch-2.8.patch
          36 kB
          Daryn Sharp
        9. HDFS-7967-branch-2.patch
          33 kB
          Daryn Sharp

        Issue Links

          depends upon

          Bug - A problem which impairs or prevents the functions of the product. HDFS-7990 IBR delete ack should not be delayed

          • Major - Major loss of function.
          • Resolved
          is related to

          Improvement - An improvement or enhancement to an existing feature or task. HDFS-9260 Improve the performance and GC friendliness of NameNode startup and full block reports

          • Major - Major loss of function.
          • Resolved
          relates to

          Improvement - An improvement or enhancement to an existing feature or task. HDFS-11310 Reduce the performance impact of the balancer (trunk port)

          • Critical - Crashes, loss of data, severe memory leak.
          • Open

          Activity

            People

              daryn Daryn Sharp
              daryn Daryn Sharp
              Votes:
              1 Vote for this issue
              Watchers:
              23 Start watching this issue

              Dates

                Created:
                Updated: