0

I have 10 Data nodes the replication factor is 3,file size is 150 and the block size is 64. So file will be splits into three blocks B1,B2,B3. So client asks Name Node for the availability of Data nodes for writing B1 block. My question is how many Data nodes will be provided to client by Name node for writing B1 block. And one more question is how the writes for B1,B2,B3 will be happend either parallelly or sequentially

sidhartha pani
  • 623
  • 2
  • 12
  • 23
  • Is there a particular problem you're encountering/trying to solve? This seems like you're just asking for background on how Hadoop works. – flyingmeatball Jan 25 '17 at 14:44

1 Answers1

2

  1. Data will be written to just one datanode by client, rest replication is taken care by the datanodes itself on namenode instruction. Replica placement: while a datanode receives data of the block from the client, the datanode saves the data in a file, which represents the block, and, simultaneously re-sends the data to another datanode, which is supposed to create another replica of the block.

  2. Blocks are copied sequentially

piyush pankaj
  • 725
  • 1
  • 12
  • 24