面向行为可信的大数据安全系统形式化描述

Formal Description of Trusted Behavior Oriented Security System for Big Data

大数据的存储和分析通常是在云计算环境下利用Hadoop计算框架进行处理,而基于云的大数据安全风险来源于云计算服务商以及远程数据应用者对未授权文件及内容的越权操作。针对云中大数据应用的安全问题,提出了面向行为可信的大数据服务系统。采用Z形式化方法对该系统进行描述,给出了读取、更新、添加及删除操作的抽象行为定义。根据模型中的抽象行为,划分可信行为与恶意行为,并给出了各个操作的安全分析。最后通过Z/EVES工具对该系统描述的正确性进行了证明和验证。

Usually the process of storing and analyzing with big data is under the framework of Hadoop within the environment of cloud computing while the risks of big data application on cloud derive from cloud service providers and remote users who access data without authorization. Based on the security issues of big data applying on cloud, a security system for big data oriented by trusted behavior was given. The system with Z formal specification was described and the definition of abstract behaviors which contains reading, updating, adding and deleting operation was given. According to the abstract behavior model, behaviors were divided into trusted behavior and malicious behavior and security analysis on each operation was given. By using Z/EVES tool verified and proved the correctness about the description.