期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2016
卷号：89
期号：1
出版社：Journal of Theoretical and Applied
摘要：While the utilization of Map Reduce methods, (for example, Hadoop) for broad data research has been normally known and investigated, we have of late seen an impact in the quantity of strategies made for thinking data giving. These more recent techniques deal with cloud OLTP programs, though they typically do not support ACID dealings. HBase is an open-source distributed NoSQL store that is commonly used by many Internet businesses to manage their big information processing programs (e.g. Face book or MySpace manages an incredible number of information each day with HBase). Optimizations that can improve the efficiency of HBase are of vital passions for big information programs that use HBase or Big Table like key-value shops. In this document we research the problems natural in mis-configuration of HBase groups, such as circumstances where the HBase standard options can lead to inadequate efficiency. We create HConfig, a semi automated settings administrator for improving HBase system efficiency from several measurements. Due to the space restriction, this document will concentrate on how to improve the efficiency of HBase information loading machine using HConfig. Through this research we believe that the significance of source flexible and amount of work aware auto-configuration management and the design concepts of HConfig. Our trial results show effective group map decreasing in information research in database integration.
关键词：Measurement; Performance; Bulk Loading; Optimization; Big Data; HBase Configuration
