基于异或的隐私保护码优化研究
发布时间：2018-12-12 20:25
【摘要】：随着存储系统的发展,为了满足当前高速增长的信息数据量对存储的需求,云存储行业迅速兴起。然而,单云存储面临着数据保密性、安全性、可用性和厂商锁定的风险。基于异或的非系统纠删码-隐私保护码(PPC)可以用来构造具有隐私保护能力的多云存储系统,从而在很大程度上解决上述问题。主要针对PPC编码算法进行优化,以提高编码运行性能。通过设计搜索PPC的最优调度来减少编码过程中的异或次数。由于PPC的编码/解码计算可以表示为生成矩阵(0/1矩阵)和数据向量的乘法,直观上计算量与生成矩阵中1的数目成正比,因此通过对计算次序的优化调度可以获得更好的性能。首先,设计并实现搜索PPC最优调度次序的算法,利用此算法寻找计算性能最优者,可优化具有隐私保护能力的多云存储系统的性能。其次,在基于最优调度次序的编码算法的基础上,利用AVX2技术的SIMD并行优化来提高编码过程中的每次异或的性能。实验表明,基于最优调度的编码性能提高了34.8%,进行SIMD并行优化后进一步提高了107.1%。
[Abstract]:With the development of storage system, cloud storage industry is rising rapidly in order to meet the demand of high speed information data volume. However, single cloud storage faces risks of data confidentiality, security, availability, and vendor locking. Non-system erasure code based on XOR-privacy protection code (PPC) can be used to construct a multi-cloud storage system with privacy protection ability to solve the above problem to a large extent. The PPC coding algorithm is optimized to improve the performance of the coding. In order to reduce the number of XOR in the coding process, the optimal scheduling of searching PPC is designed. Because the encoding / decoding calculation of PPC can be expressed as the multiplication of the generating matrix (0 / 1 matrix) and the data vector, the computation amount is directly proportional to the number of 1 in the generated matrix. Therefore, better performance can be obtained by optimizing the order of calculation. Firstly, an algorithm to search for the optimal scheduling order of PPC is designed and implemented. The algorithm can be used to find the optimal computing performance, which can optimize the performance of the multi-cloud storage system with privacy protection. Secondly, on the basis of the coding algorithm based on the optimal scheduling order, SIMD parallel optimization based on AVX2 technology is used to improve the performance of each XOR in the coding process. Experiments show that the coding performance based on optimal scheduling is improved by 34.8. after parallel optimization of SIMD, it is further improved by 107.1.
【作者单位】： 南开大学计算机与控制工程学院;
【分类号】：TP309;TP333
本文编号：2375196
[Abstract]:With the development of storage system, cloud storage industry is rising rapidly in order to meet the demand of high speed information data volume. However, single cloud storage faces risks of data confidentiality, security, availability, and vendor locking. Non-system erasure code based on XOR-privacy protection code (PPC) can be used to construct a multi-cloud storage system with privacy protection ability to solve the above problem to a large extent. The PPC coding algorithm is optimized to improve the performance of the coding. In order to reduce the number of XOR in the coding process, the optimal scheduling of searching PPC is designed. Because the encoding / decoding calculation of PPC can be expressed as the multiplication of the generating matrix (0 / 1 matrix) and the data vector, the computation amount is directly proportional to the number of 1 in the generated matrix. Therefore, better performance can be obtained by optimizing the order of calculation. Firstly, an algorithm to search for the optimal scheduling order of PPC is designed and implemented. The algorithm can be used to find the optimal computing performance, which can optimize the performance of the multi-cloud storage system with privacy protection. Secondly, on the basis of the coding algorithm based on the optimal scheduling order, SIMD parallel optimization based on AVX2 technology is used to improve the performance of each XOR in the coding process. Experiments show that the coding performance based on optimal scheduling is improved by 34.8. after parallel optimization of SIMD, it is further improved by 107.1.
【作者单位】： 南开大学计算机与控制工程学院;
【分类号】：TP309;TP333
【相似文献】
相关期刊论文 前3条
1 龚庆悦;;计算机和网络之隐私问题探析[J];医学信息学杂志;2009年08期
2 康琳;李秀华;王卫东;;参与感知世界的激励机制研究[J];单片机与嵌入式系统应用;2013年08期
3 ;[J];;年期
相关重要报纸文章 前3条
1 中国科学院信息安全国家重点实验室 左晓栋;P3P技术，网络隐私的保护神[N];电脑报;2001年
2 ;RFID蠢蠢欲动[N];网络世界;2003年
3 陈翔 吴锋;谁偷窥了你的隐私[N];中国计算机报;2006年
相关硕士学位论文 前3条
1 姚川;基于分布式云存储的隐私保护协议研究[D];福建师范大学;2015年
2 杨亭亭;面向云存储的访问控制方法研究[D];华北电力大学(北京);2016年
3 周静岚;云存储数据隐私保护机制的研究[D];南京邮电大学;2014年
，本文编号：2375196
本文链接：https://www.wllwen.com/kejilunwen/jisuanjikexuelunwen/2375196.html
