Auto-Scale Multi-Node Hadoop Cluster – An Experimental Setup
Praseetha V. M
, Pratik Sen , S. Vadivel
Cluster Computing, Hadoop, Distributed processing, Cluster
With the plateauing of computational hardware, the need to get the maximum performance both reliably and at low cost has seen to the revival of cluster computing. Cluster computing makes use of a large number of low cost ordinary everyday computers. Apache Hadoop is a scalable, flexible, cost effective and fault tolerant open source software system framework which enables distributed processing across clusters. As Hadoop is statically set up all configurations are defined before deploying the actual nodes. In such a case mirror images of the hardware partitions are used and then applied to the target machines. However such a case causes a very brittle set up which can cause a catastrophic failure with the failure of the master nodes. To reinstate the nodes the entire cluster would generally need to be brought down. To combat this weakness we successfully set up a novel multi node approach with two nodes and is presented in this paper.
"Auto-Scale Multi-Node Hadoop Cluster – An Experimental Setup", IJSDR - International Journal of Scientific Development and Research (www.IJSDR.org), ISSN:2455-2631, Vol.1, Issue 8, page no.123 - 132, August-2016, Available :https://ijsdr.org/papers/IJSDR1608016.pdf
Volume 1
Issue 8,
August-2016
Pages : 123 - 132
Paper Reg. ID: IJSDR_160671
Published Paper Id: IJSDR1608016
Downloads: 000347231
Research Area: Engineering
Country: Kottayam, Kerala, India
ISSN: 2455-2631 | IMPACT FACTOR: 9.15 Calculated By Google Scholar | ESTD YEAR: 2016
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.15 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: IJSDR(IJ Publication) Janvi Wave
