Elsevier

Information Sciences

Volume 465, October 2018, Pages 219-231
Information Sciences

Block-secure: Blockchain based scheme for secure P2P cloud storage

https://doi.org/10.1016/j.ins.2018.06.071Get rights and content

Abstract

With the development of Internet technology, the volume of data is increasing tremendously. To tackle with large-scale data, more and more applications choose to enlarge the storage capacity of users’ terminals with the help of cloud platforms. Before storing data to an untrusted cloud server, some measures should be adopted to guarantee the data security. However, the communication overhead will increase dramatically when users transmit files encrypted by a traditional encryption scheme. In this paper, we address the above problems by proposing a blockchain-based security architecture for distributed cloud storage, where users can divide their own files into encrypted data chunks, and upload those data chunks randomly into the P2P network nodes that provide free storage capacity. We customize a genetic algorithm to solve the file block replica placement problem between multiple users and multiple data centers in the distributed cloud storage environment. Numerical results show that the proposed architecture outperforms the traditional cloud storage architectures in terms of file security and network transmission delay. On average, the file loss rate based on the simulation assumptions utilized in this paper is close to 0% on our architecture while it’s nearly 100% and 71.66% on the architecture with single data center and the distributed architecture using genetic algorithm. Besides, with proposed scheme, the transmission delay on the proposed architecture is reduced by 39.28% and 76.47% on average on the user’s number and the number of file block replicas, respectively, in comparison to the architecture with single data center. Meanwhile, the transmission delay of file block replicas is also reduced by 41.36% on average than that on the distributed architecture using genetic algorithm.

Introduction

Cloud storage is a kind of system with distributed data centers that takes advantage of virtualization technology and provides interfaces for data storage. It also makes servers or data centers be able to work together for conveniently sharing and accessing resources. Recently, cloud storage has received massive attractions in personal and business organizations because it’s convenient and efficient.

In order to access application resources from anywhere and at any time, users have to move their data into the cloud. Cloud provides benefits like flexibility, automatic software update, disaster tolerance, cost reduction and etc. For advantages, challenges and key technologies in different types of cloud storage one can refer to Wang et al. [40] and Li et al. [12].

It’s important to protect users’ privacy [46] and data security [47] because the data may leak when users store their data in the cloud. Meanwhile, the influence of cloud storage security is widely and there is an increasing public concern about users’ privacy [35], [36]. However, there still exists weakness for the existing cloud storage architectures, such as centralized data storage that severely harms server physical security and the need for trusted third-party which are nightmares for the privacy of users’ data.

For the current distributed cloud storage, the data stored in several data centers are not fully distributed. The data are still stored in several data centers at high density, and a massive amount of data will be leaked even if one of the data centers was broken down. Public media all around the world have repeatedly reported security issues related to cloud storage in recent years, such as users’ privacy file leakage on iCloud [25]. Unfortunately, there are still no effective solutions for the security of distributed cloud storage.

Zyskind et al. propose an architecture and their architecture uses blockchain to protect personal data through distributed storing file access permissions in the blockchain, but its data storage still uses a centralized cloud and requires a trusted third-party to support [50]. Blockchain is a distributed database system, it also can be regarded as a number of nodes jointly maintained by the Distributed Ledger Technology (DLT), which is difficult to tamper, forge and trace [29]. The blockchain records all the information of the transactions, and once the data enter the blockchain, almost nobody can change it. This unchangeable feature is not derived from the use of a certain operation but from the blockchain system and the mechanism itself. This makes the use of blockchain technology easier and more secure than other security technologies. For example, the work [26] introduces how to use blockchain technology in intrusion detection systems (IDSs) and Lin et al. [19] utilizes blockchain to protect the security of users’ data.

As its typical feature of blockchain, which includes peer-to-peer (P2P) communication, we utilize unstructured P2P network in our distributed storage architecture. A P2P network shares resources among nodes rather than concentrating them in a single data center or server [31]. In a P2P system, nodes are pooled together to provide their network resources [34]. Therefore, we can put all users’ vacant storage space as a storage pool providing a cloud storage service to other users.

In this new cloud storage architecture, we need to solve the following research challenges. (i) A novel storage strategy is necessary because existing distributed cloud storage strategy stores data centrally in several data centers and once a data center is broken down, a large amount of data may leak. (ii) In order to compare the network performance, a heuristic optimization algorithm should be utilized to optimize the NP-hard problem of the files transmission delay between users and data centers. (iii) Storing file blocks randomly into the nearest nodes in the network that have vacant storage capacity is an intuitive challenge because this process will consume a massive amount of computation and communication resources. Especially in the large-scale network, where the computational complexity will grow exponentially.

To deal with the disadvantages and challenges as mentioned above, we integrate distributed cloud storage and blockchain technology to propose a blockchain-based distributed cloud storage architecture that can provide secure and reliable cloud storage services for enterprises or individual users. In this work, related works on cloud storage are first to introduce, then a blockchain-based distributed cloud storage architecture and its security analysis are proposed and finally, three cloud storage architectures including a centralized cloud storage architecture and a distributed cloud storage architecture are compared through simulations. To the best of our knowledge, this paper is the first work for the blockchain-based architecture with completely distributed cloud storage.

This paper is distinct from our previous work [16] in many aspects. First, Li et al. [16] has been significantly extended by adding investigations for the further security analysis and the analysis of replica number in all three architectures. Moreover, the numerical results of further security analysis and network performance analysis on replica number are provided in this paper. The main contributions of this paper are summarized as follows.

  • We propose a blockchain-based distributed cloud storage architecture to provide more secure and reliable cloud storage services for enterprises or individual users.

  • We customize a genetic algorithm to solve the file block replica placement problem between multiple users and multiple data centers in distributed cloud storage environment.

  • We conduct simulations on the proposed architecture for security and network performance, in comparison with other two different cloud storage architectures.

  • On average, the file loss rate in this work is almost 0% based on the assumptions made in the simulation for our architecture while it’s nearly 100% and 71.66% for the other two architectures.

  • The transmission delay on the proposed architecture is averagely reduced by 39.28% and 76.47% on the user’s number and the number of file block replicas, respectively, in comparison to the architecture with single data center.

The rest of this paper is organized as follows. In Section 2, we discuss related work of cloud storage and blockchain technology. Then we propose a novel blockchain based distributed cloud storage architecture in Section 3. The network performance and security analysis are presented both for the proposed architecture and for the traditional cloud storage architecture in Section 4. Numerical simulations are presented in Section 5. We conclude this paper in Section 6.

Section snippets

Related works

Cloud storage is a kind of Internet technology for sharing resources with IT-related capabilities and it is important to either enterprises or individual users. Traditional security strategies mainly focus on information encryption [18], [21], [45], data deduplication [13], [14], [15], access control [1], [12], [33], [37], privacy-preserving keyword search [10], [17], [23], [39], network performance improvement [22] and etc. Recently, application data are becoming more and more intensive and a

Architecture design

In this section, we present a blockchain-based security architecture for distributed cloud storage. As can be seen in Fig. 1, in this architecture, we first divide users’ files into several blocks with the same size, encrypt these file blocks, sign them through a Digital Signature Algorithm (DSA) and upload them to a P2P network. Then we utilize blockchain technology as a trading mechanism between users who need cloud storage service and users who supply their vacant storage space. Furthermore,

Network performance & security analysis

In this section, we provide security analysis by comparing three kinds of cloud storage architectures (subsection A, B, and C) during the random process of users store and retrieve their files in the cloud. Our target aims to reduce network latency and maximum file loss rate of security events in the architectures. As for network performance, we consider transmission time as a criterion. Meanwhile, since the research works of consistency are almost based on the assumption that channels are

Experiment & analysis

In this section, we evaluate the performance of our architecture with extensive numerical analysis. Evaluation methods have been proposed in Section 4, and the evaluations simulate the processes of users store and retrieve their files in the cloud. In Section 5.1 we introduce the environment setup and parameter setting, and in Section 5.2 we analyze the results of these evaluations.

Conclusion

This paper has proposed a blockchain-based security architecture for distributed cloud storage. The proposed architecture has been compared with other two traditional architectures in terms of security and network transmission delay. Based on the simulation assumptions utilized in this paper, the file loss rate of the proposed architecture outperforms other two traditional architectures on average. Meanwhile, the network performance of the traditional distributed architecture has been improved

Acknowledgment

Part of the work was presented in IEEE International Symposium on Parallel and Distributed Processing with Application (ISPA) (2017). This work was supported by the National Key R&D Program of China (2018YFB1003201), the National Natural Science Foundation of China (61672171, 61702115, 61702114), Major Research Project of Educational Commission of Guangdong Province (2016KZDXM052), Guangdong Natural Science Foundation (2018B030311007), Opening Project of Guangdong Province Key Laboratory of Big

References (50)

  • W. Diffie et al.

    New directions in cryptography

    IEEE Trans. Inf. Theory

    (1976)
  • H.G. Do et al.

    Blockchain-based system for secure data storage with private keyword search

    Services (SERVICES), 2017 IEEE World Congress on

    (2017)
  • S. Gueron et al.

    Sha-512/256

    Eighth International Conference on Information Technology: New Generations

    (2011)
  • D. Johnson et al.

    The elliptic curve digital signature algorithm (ecdsa)

    Int. J. Inf. Secur.

    (2001)
  • M. Kiran, M. Stanett, Bitcoin risk analysis, 2015,...
  • N. Koblitz

    Elliptic curve cryptosystems

    Math. Comput.

    (1987)
  • B. Li et al.

    Hybridoram: practical oblivious cloud storage with constant bandwidth

    Inf. Sci.

    (2018)
  • H. Li et al.

    An efficient merkle-tree-based authentication scheme for smart grid

    IEEE Syst. J.

    (2014)
  • J. Li et al.

    Multi-authority fine-grained access control with accountability and its application in cloud

    J. Netw. Comput. Appl.

    (2018)
  • J. Li et al.

    Secure auditing and deduplicating data in cloud

    IEEE Trans. Comput.

    (2016)
  • J. Li et al.

    A hybrid cloud approach for secure authorized deduplication

    IEEE Trans. Parallel Distrib. Syst.

    (2015)
  • J. Li et al.

    Blockchain-based security architecture for distributed cloud storage

    Ubiquitous Computing and Communications (ISPA/IUCC), 2017 IEEE International Symposium on Parallel and Distributed Processing with Applications and 2017 IEEE International Conference on

    (2017)
  • T. Li et al.

    Differentially private naive bayes learning over multiple data sources

    Inf. Sci.

    (2018)
  • Q. Lin et al.

    An id-based linearly homomorphic signature scheme and its application in blockchain

    IEEE Access

    (2018)
  • B. Liu et al.

    Blockchain based data integrity service framework for iot data

    Web Services (ICWS), 2017 IEEE International Conference on

    (2017)
  • Cited by (0)

    View full text