Improved fault tolerance architecture for HDFS using Distributed namespaces

  Karishma Kharde,  Priyanka Shelar,  Smruti Sakhare

Interest and Knowledge about the computers and online world is growing along with the need for it. This interest is, no doubt due to continued expansion of internet. This paper is aimed to improve the fault tolerance for HDFS using Distributed Namespace. A center Piece of an HDFS file system, Namespace keeps the directory tree of all files in the file system, also tracks where across the cluster the file data is kept. A client can communicate with the namenode whenever they wish to locate or get a file mapping to the Blocks on Datanodes. A namenode is a Single Point of Failure (SPOF) for the HDFS cluster, when the namenode goes down the file system goes offline there can be an optional name node that can be hosted on a separate machine. Here present multiple namenode with Distributed namespace using Chord Protocol. Multiple namenodes will be arranged in chord ring to provide fault tolerant architecture in HDFS.

Keywords- Chord protocol, HDFS, Namespace, fault tolerant, SPOF

Unique Identification Number - IJEDR1802071
Page Number(s) - 416-419
Pubished in - Volume 6 | Issue 2 | April 2018
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Karishma Kharde,  Priyanka Shelar,  Smruti Sakhare,   "Improved fault tolerance architecture for HDFS using Distributed namespaces", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.6, Issue 2, pp.416-419, April 2018, Available at :http://www.ijedr.org/papers/IJEDR1802071.pdf