This topic contains 1 reply, has 2 voices, and was last updated by  Sankhabrata Burman 4 months, 3 weeks ago.

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
  • #3077 Reply


    The Name node allocates block to different data nodes based on 3 criteria, that are:

    1> Location of the client and the data node.
    2> Network Traffic in that area /data node, if more then blocks are given to the next nearest data node.
    3> Data Redundancy.

    The above criteria is on priority basis, i.e. from high to Low.

    #5596 Reply

    Sankhabrata Burman

    -Rack Awarness and Bandwith:
    Large HDFS instances run on a cluster of computers that commonly spread across many racks.
    In most cases, network bandwidth between machines in the same rack is greater than network bandwidth between machines
    in different racks.A simple but non-optimal policy is to place replicas on unique racks. This prevents losing data when an
    entire rack fails and allows use of bandwidth from multiple racks when reading data. This policy evenly distributes replicas
    in the cluster which makes it easy to balance load on component failure.
    However, this policy increases the cost of writes because a write needs to transfer blocks to multiple racks.

    -Proximity of DataNode:
    To minimize global bandwidth consumption and read latency, HDFS tries to satisfy the block allocation or read request
    from a replica that is closest to the reader.

Viewing 2 posts - 1 through 2 (of 2 total)
Reply To: Algorithm for Name Node to allocate block on a different Data Node?
Your information:


Your Name (required)

Your Email (required)


Phone No

Your Message


  • No products in the cart.