面向大数据的海云数据系统关键技术研究

由于数据产生成本的急速下降,人类社会产生的数据不仅以指数级别增长,而且数据的结构变得日趋复杂,使得传统的数据分析技术遇到了极大的挑战.如何对大量复杂数据进行分析和挖掘,从中提取有价值的知识用于决策,已经成为产业界和学术界的广泛关注问题,在一些国家已上升到国家战略层面.本文介绍了大数据的基本概念、特征和面临的科学问题,总结了中国科学院战略性先导科技专项“面向感知中国的新一代信息技术研究”中“海云数据系统关键技术研究与系统研制”课题的一些先期成果,为开发大数据管理、分析和挖掘系统提供一些参考依据.

Developing Sea Cloud Data System Key Technologies for Large Data Analysis and Mining

As the cost of data generation is going down, the data generated in the society increases exponentially. Furthermore, the complexity of the data structure is increasing as well. This phenomenon presents an extremely big challenge to the traditional data analysis techniques. How to analyze and mine big data and extract valuable knowledge for decision making has become a popular big concern in academics and industry. Some countries have raised this issue to the national strategic level. In this paper, we first introduce some basic concepts about big data, its characteristics and scientific research problems, We then summarize some initial results of the project of “research and development of key technologies for sea and cloud data systems” in the program of “sensing China oriented next generation information technologies” being carried out in Chinese Academy of Sciences. We propose further innovations on the related technologies to spearhead the research on big data management, analysis and mining in the near future.