In HDFS, the Namenode handles all the metadata about the files. Client contacts the Namenode for accessing any file to get its mapping to the blocks on Datanodes. Namenode has to store this data in its RAM to allow faster access to clients on their requests. Hence HDFS Datanodes’ metadata is restricted by the capacity of the RAM. This makes the Namenode as the single point of failure in HDFS. In our approach we present multiple Namenodes with distributed namespace using Chord protocol. Multiple Namenodes will be arranged in Chord ring to provide scalable and fault tolerant architecture in HDFS.
