摘要
全面应对大数据的挑战需要来自存储技术、下一代网络、处理器、计算模型等各个领域的创新,粒计算是在求解问题过程中使用"粒"的理论、方法、技术和工具的集合,适用于近似求解有不确定性和层次结构的问题。文章综述了大数据处理的研究现状,分析了当前大数据处理研究存在的局限性,根据运用粒计算方法解决问题的不同特征,归纳了粒计算的3种基本模式,回顾了各种模式的相关研究工作,讨论了粒计算应用于大数据处理的可行性与优势,并探讨了在大数据的粒计算处理框架中需要解决的各个关键问题。
A comprehensive response to the challenges of big data requires innovations in all areas of storage technology, next generation network, processor, computational models, etc. The granular computing is the collection of theory, methods, techniques and tools that use "kernels" in solving problems, and is suitable for solving the problems with approximate the uncertainty and hierarchical structure. This paper summarizes the present research situation of big data, analyzes the limitations of current big data processing research. According to the different characteristics of granular computing method used to solve the problem, summarizes the three basic mode of granular computing, reviews the related research work of various modes, discusses the feasibility and advantages of application of granular computing in large data processing, and discusses the the key problems need to be solved in the framework of granular computing processing of big data.
出处
《无线互联科技》
2018年第1期75-76,共2页
Wireless Internet Technology
关键词
大数据
粒计算
数据信息
big data
granular computing
data information