基于双哈希模糊布隆滤波器云存储数据融合  

Cloud storage design based on fusion of double hash fuzzy cross Bloom filter

作　　者：洪文圳[1] 李冬睿[1] 沈阳[2] HONG Wen-zhen;LI Dong-rui;SHEN Yang(School of Computing,Guangdong Agricultural Industry and Business Polytechnic College,Guangzhou 510507,China;School of Software,South China University of Technology,Guangzhou 510641,China)

机构地区：[1]广东农工商职业技术学院计算机学院,广东广州510507 [2]华南理工大学软件学院,广东广州510641

出　　处：《计算机工程与设计》2021年第8期2152-2158,共7页Computer Engineering and Design

基　　金：广东省教育技术中心信息化应用融合创新课题基金项目(19JX06244)。

摘　　要：为提高跨多个区域大数据存储效率,提出一种基于布隆(Bloom)滤波器(BF)的海量数据存储空间部署策略。采用模糊交叉方法(FFBF),使用模糊交叉操作合并压缩两个Bloom滤波器,实现散列数据在两个Bloom滤波器的共享容纳,减少海量数据存储需求;利用双哈希计算k个哈希函数降低计算成本。实验结果表明,所提算法的误报受压缩操作(即失效数据)的影响很小,数据衰减缓慢,允许流数据在内存中驻留相当长的时间。To improve the efficiency of large data storage across multiple regions,a deployment strategy of massive data storage space based on Bloom filter(BF)was proposed.The fuzzy crossover method(FFBF)was used to compress the two Bloom filters by combining the fuzzy crossover operation,so as to realize the sharing of hash data in the two Bloom filters and reduce the sto-rage requirements of massive data.Double hash was used to calculate k hash functions to reduce the cost.Experimental results show that the false positives of the proposed algorithm are barely affected by the compression operation(i.e.invalid data),the data decay is slow,and the stream data are allowed to stay in the memory for a long time.

关 键 词：双哈希函数 模糊交叉 BLOOM滤波器 融合 云存储 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]