Secure Dynamic Big Graph Data: Scalable, Low-Cost Remote Data Integrity Checking

Secure Dynamic Big Graph Data: Scalable, Low-Cost Remote Data Integrity Checking
复制标题

DOI：
10.1109/access.2019.2892442
复制
发表时间：
2019
期刊：
IEEE Access
影响因子：
3.9
通讯作者：
Yu Lu;F. Hu
Yu Lu;F. Hu
中科院分区：
计算机科学3区
文献类型：
--
作者：
Yu Lu;F. Hu

文献摘要

相似文献

大图作为一种特殊类型的大数据，有着广泛的应用。远程数据完整性检查（RDIC）方案使公共云能够有效地让客户相信他们的图数据存储正确，而无需检索实际数据内容。现有方案通常通过采用认证数据结构（ADS）来支持可验证更新，例如默克尔哈希树（MHT）。现有的RDIC方案应用于大图的主要障碍是缺乏对可验证的高效率子图操作的支持。在本文中，我们提出了一种新的大图 RDIC 方案，称为 GVD-RDIC，以支持高效的公共审计和可验证的动态图更新。我们设计了基于图 Voronoi 图 (GVD) 的新型 ADS 和增强的 MHT，以解决图结构的完整性和可验证的子图更新。此外，我们的 ADS 可以应用于任何类型的图表。此外，我们的GVD-RDIC方案采用了同态验证器的新结构，以便为公共审计师提供索引验证。使用未挑战的块构造的服务器响应将被拒绝。所提出的方案在随机预言模型下被证明是安全的。理论分析和仿真结果都表明我们的方案对于现实世界的大图是可行的。
As a special type of big data, the big graph has wide applications. The remote data integrity checking (RDIC) scheme enables public clouds to efficiently convince the clients that their graph data are stored properly, without the need of retrieving the actual data contents. The existing schemes support verifiable update often by adopting the authentication data structures (ADSs), e.g., Merkel hash tree (MHT). The main obstacle for applying the existing RDIC schemes to big graphs is due to the lack of support for verifiable sub-graph operations with high efficiency. In this paper, we propose a new RDIC scheme for big graphs, called GVD-RDIC, to support public auditing and verifiable dynamic graph updates with high efficiency. We have designed novel ADS based on the graph Voronoi diagram (GVD) and enhanced MHT to address the integrity of graph structure and verifiable sub-graph updates. In addition, our ADS can be applied to any type of graphs. Moreover, our GVD-RDIC scheme adopts a new construction for the homomorphic authenticator to enable index verifications for public auditors. A server response constructed with unchallenged block will be rejected. The proposed scheme is proven secure under a random oracle model. Both the theoretical analysis and the simulation results show that our scheme is practicable for real-world big graphs.