COMPARATIVE STUDY OF MACHINE LEARNING TOOLS HADOOP DISTRIBUTED FILE SYSTEM, CASSANDRA FILE SYSTEM, QUANT CAST FILE SYSTEM TO ENHANCE THE EFFICACY OF DATA ANALYTICS ON UNSTRUCTURED DATA

Mukul Ganghas

Vol. 5, Jan-Dec 2019

Page Number: 134 - 142

Abstract:

Objective: With the emergence of the belief of the "Internet of Things (IoT)," an enormous quantity of Data is being generated through the sensors and other computing gadgets and chips. This paper is an try to provide a lucid contrast among 3 outstanding technologies used for managing Big Data, viz. HDFS, Cassandra file system, and Quant robust record gadget. Apart from these three ultimate report systems, the paper additionally explores a newly proposed A Train Distributed System for dealing with Big Data. Methods: An internal perspective of the above-stated record systems in details thinking about diverse components for coping with massive information has been described. The paper also presents sagacity on the conditions in which these technologies are useful. Findings: Effective tackling of the 5 V's (Variety, Volume, Velocity, Veracity, and Value) of Big Data has to turn out to be a hard assignment for the researcher around the sector. Hadoop is one such generation that's open supply and is capable of coping with extensive records powerfully. It breaks the huge statistics into fixed-sized chunks referred to as a block, and these blocks are saved at awesome places in a distributed manner. The Cassandra document gadget is an alternative to Hadoop, which eliminates the single factor failure hassle of Hadoop as it follows master-less peer to look distributed ring architecture instead of customer server architecture. The 0.33 era is the quant forged file device that's written in the C++ language. It likewise handles the large statistics powerfully and efficiently. Moreover, it claims to keep as much as fifty per cent of the disk space by using imposing erasure encoding. Application: The concerned agency to apply any of these to be had frameworks for coping with large facts relying upon their nature of wishes

References

Back Download