On the Research of Big Data Storage

Abstract

This paper want to research the storage of big data, the paper from the processing mode of big data start, study big data process, requirements, management and storagea. Analysis currently being studied big data storage technology GFS, Bigtable, HDFS, NoSQL, multi-Replication and disaster recovery technology. The aim is for the storage of big data parepare. KEYWORD: Big data; big data storage; GFS; Bigtable; HDFS; NoSQL International Conference on Industrial Technology and Management Science (ITMS 2015) © 2015. The authors Published by Atlantis Press 1410 the additional 0.5 times storage for intermediate result set processing and storage[2]. Big data is valuable, but to mining the value of big data is similar to the waves in panning. This value not once discovery, it requires a continuous change process, such as video monitoring, producing 24 hours of video every day, most of the data is no value, may be a few seconds to capture shots to a certain physical characteristics of offenders, is precious to the public security sector, for those few seconds, 24 hours all data must be save. This is a typical feature of the big data, low density high value. The store-then-prcess batch processing is more in line with big data processing requirement. Big data will become part of the modern social infrastructure gradually, like roads, railways, ports, utilities and communications networks as indispensable. Big data is not only because of possession and exhausted, but also with the use and dissemination of the continuous enrichment and growth. As you can imagine, the data will be the future of DNA. But having the data volume, quality and application will form a significant segment of data resources and competitive advantage. Therefore, big data will become the countries for the next strategic frontier. With the rise of smart economy, collecting data, master data, and use data will become the country's core competitiveness of enterprises. According to the big data storage mode, the data and information always back and forth on the layer of processing and storage converted, get the information from the data, the information can also be obtained from the new data. The converion figure as follows figure2.