Distributed Data Storage Systems for Data Survivability in Wireless Sensor Networks using Decentralized Erasure Codes

doi:10.1016/j.comnet.2016.01.008

Computer Networks

Volume 97, 14 March 2016, Pages 113-127

https://doi.org/10.1016/j.comnet.2016.01.008 Get rights and content

Abstract

Achieving reliability in Wireless Sensor Networks (WSNs) is challenging due to the limited resources available. In this study, we investigate the design of data survivability schemes using decentralized storage systems in WSNs. We propose a data storage system design based on Decentralized Erasure Codes (DEC) that features a simple and decentralized construction of the target code. The proposed framework allows sensor nodes to cooperate to build an erasure code-based storage that can tolerate a given failure/erasure rate. Code construction and decoding can both be performed randomly allowing for a distributed operation with no prior setup or coordination between source nodes. Further, we present two approaches that utilize Random Linear Network Coding (RLNC) to enhance the proposed scheme in order to achieve energy efficiency. We present the theoretical basis of the schemes then validate and evaluate their performance through simulations.

Introduction

Wireless Sensor Network (WSN) technology is being increasingly deployed in a diverse range of applications. Intelligent Transportation Systems (ITSs) [1], Smart Grids [2], and the Internet of Things (IoT) [3] are just a few examples of technologies where WSNs are used. Generally, WSNs are comprised of sensor nodes that are equipped with one or multiple sensors, a processing unit, and a wireless communication module. Sensor nodes cooperate in monitoring a phenomenon of interest and in relaying the sensed data to a sink node for processing. When produced in large numbers, sensor nodes can be extremely inexpensive, and hence they can be deployed in greater numbers to build large scale networks. WSNs have stringent constraints, especially regarding power consumption and scalability. Furthermore, reliability becomes a key requirement for WSNs when deployed in unattended applications or under harsh working conditions.

To preserve the sensed data captured by sensor nodes, WSNs nodes can benefit from using Distributed Data Storage Systems (DDSSs) technology. Data storage systems represent an essential component of today’s networks and they have been researched for a long time. Lately, data storage technology is being revisited especially in the contexts of Content Centric Networking (CCN) [4] and cloud computing [2]. DDSSs utilize hardware redundancy and data replication to protect data in case of possible failures. More specifically, given a data packet, a DDSS replicates the packet over multiple physical storage devices, such that when a subset of these devices fails, the data packet can be retrieved from the surviving ones.

In this study, our goal is to design a DDSS that is tailored for WSNs data reliability applications. For that, we first introduce the notion of data survivability as a quantitative parameter that links the amount of redundancy required to the maximum failure that can be tolerated. We then show how data survivability can be useful by implementing a data survivability scheme, called Decentralized 30 Erasure Codes for Data Survivability (DEC-DS). DEC-DS is based on Decentralized Erasure Codes (DEC) [5], [6], [7]. Besides being decentralized, DEC has a predictable algebraic structure allowing for quantifiable performance. After that, we present two methods to enhance the energy efficiency of DEC-DS by exploiting Network Coding (NC). The two schemes are referred to as DEC Encode-and-Forward (DEC-EaF) and DEC Encode-and-Disseminate (DEC-EaD). NC [8] has emerged as an information-theoretic tool and has been shown to decrease energy consumption and complexity while increasing throughput and reliability [9]. Random Linear Network Coding (RLNC) [10] has been later proposed as a practical implementation of Network Coding. In this study, we utilize RLNC to increase the efficiency of the proposed storage system by reducing communication overhead and consequently energy requirements. The main contributions of this paper are introducing the notion of data survivability and presenting the three data storage schemes, DEC-DS, DEC-EaF, and DEC-EaD.

The remainder of the paper is organized as follows. In Section 2, we present some background material and review related work. The proposed data survivability framework is discussed in Section 3. Section 4 shows two schemes using RLNC to improve the efficiency of the proposed data survivability application. Experiments and results are discussed in Section 5. Finally, Section 6 concludes the paper. Some important results from the theory of random matrices over finite fields, which will be used in designing the codes, are presented in Appendix A.

Section snippets

Background and related work

Before we discuss the proposed schemes, we present the advantages and disadvantages of replication and encoding-based storage. We then present the concept of data survivability and how it differs from network survivability. We also present an overview of Fountain Codes and DEC; and survey related literature on DDSSs in WSNs.

Decentralized Erasure Codes for Data Survivability (DEC-DS)

The design of an erasure code in a centralized setup is quite different than that in a decentralized one. The difference is illustrated in Fig. 2. In a centralized code (Fig. 2(a)), all k native data blocks are available at a single encoder. Hence, when sampling a degree d from the degree distribution ρ(k), d can be exactly matched since all k native packets are available to the encoder. In other words, if the degree sampled from the distribution is d, the encoder has all the k data packets

Routing and energy efficiency

By restricting their role to pure routing/forwarding, relay nodes are not fully utilized by existing DEC schemes when disseminating data. So, following generating a source packet, choosing a set of candidate storage nodes, and forwarding the packet by source nodes, relay nodes help forward data packets to storage nodes without manipulating them. However, based on the argument that the cost of communication is generally much higher than processing on wireless nodes, we utilize the coding

Performance evaluation

As we stated earlier, our proposed schemes target resource limited networks. Therefore, they must achieve data survivability at a reasonable energy cost. We also conjectured that data survivability can be achieved using less redundancy than required by the original DEC. In this section we show through experimentation that DEC-DS can guarantee data survivability while reducing redundancy requirements. Note that redundancy reduction impacts the energy required for data dissemination and encoding.

Conclusion

In this paper, we introduce a DDSS for data survivability in WSNs based on DEC. The framework aims at determining the amount of redundancy in both storage and data, and the maximum level of failure that can be tolerated without losing in-network data. Compared to the original DEC which aim at reducing DO, the proposed schemes target achieving data survivability. Due to the random nature of the DEC-DS scheme, it can be implemented in a decentralized manner without coordination between the sensor

Louai Al-Awami is currently an Assistant Professor at the Computer Engineering Department, at King Fahd University of Petroleum and Minerals (KFUPM), Saudi Arabia. He earned his BSc and MSc in Computer Engineering from the Computer Engineering, at KFUPM, in 2002 and 2006, and his Ph.D. from the Electrical and Computer Engineering Department at Queen’s University, Canada, respectively. He has also taught many courses and laboratories while working as a lecturer at KFUPM. His research interest

References (39)

Q. Wang et al.
Dependable and secure sensor data storage with dynamic integrity assurance
ACM Trans. Sen. Netw.
(2011)
S. Katti et al.
Xors in the air: Practical wireless network coding
IEEE/ACM Trans. Netw.
(2008)
W. Wu et al.
Distributed mutual exclusion algorithms for intersection traffic control
IEEE Trans. Parallel Distrib. Syst.
(2015)
S. Bera et al.
Cloud computing applications for smart grid: A survey
IEEE Trans. Parallel Distrib. Syst.
(2015)
H. Ning et al.
Aggregated-proof based hierarchical authentication scheme for the Internet of Things
IEEE Trans. Parallel Distrib. Syst.
(2015)
Z. Ren et al.
CCN-WSN - A lightweight, flexible content-centric networking protocol for wireless sensor networks
Proceedings of the 2013 IEEE Eighth International Conference on Intelligent Sensors, Sensor Networks and Information Processing
(2013)
A.G. Dimakis et al.
Ubiquitous access to distributed data in large-scale sensor networks through decentralized erasure codes
Proceedings of the Fourth International Symposium on Information Processing in Sensor Networks (IPSN ’05), Piscataway, NJ, USA
(2005)
A. Dimakis et al.
Distributed data storage in sensor networks using decentralized erasure codes
Proceedings of the Thirty-eighth Asilomar Conference on Signals, Systems, and Computers
(2004)
A. Dimakis et al.
Decentralized erasure codes for distributed networked storage
IEEE Trans. Inf. Theory
(2006)
R. Ahlswede et al.
Network information flow
IEEE Trans. Inf. Theory
(2000)

C. Fragouli et al.

Network coding: An instant primer

ACM SIGCOMM Comput. Commun. Rev.

(2006)

S.-Y.R. Li et al.

Linear network coding

IEEE Trans. Inf. Theory

(2003)

H. Weatherspoon et al.

Erasure coding vs. replication: A quantitative comparison

Proceedings of the First International Workshop on Peer-to-Peer Systems (IPTPS ’01)

(2002)

S. Acedan’ski et al.

How good is random linear coding based distributed networked storage

Proceedings of the First Workshop on Network Coding, Theory, and Applications (NetCod 2005)

(2005)

F.A. Kuipers

An overview of algorithms for network survivability

ISRN Commun. Netw.

(2012)

J. Xu et al.

Enhancing survivability in virtualized data centers: A service-aware approach

IEEE J. Sel. Areas Commun.

(2013)

J.W. Byers et al.

A digital fountain approach to reliable distribution of bulk data

Proceedings of the ACM Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication (SIGCOMM ’98)

(1998)

I.S. Reed et al.

Polynomial codes over certain finite fields

J. Soc. Ind. Appl. Math.

(1960)

M. Luby

LT codes

Proceedings of the Forty-third IEEE Annual Symposium on Foundations of Computer Science

(2002)

Cited by (20)

Edge deduplication for LoRaWAN using network coding
2023, Internet of Things (Netherlands)
LoRaWAN is a promising access technology for LPWAN IoT applications. However, the scalability of LoRaWAN has been debated due to its ALOHA-like media access mechanism. While using multiple gateways in the network can improve scalability, it may result in huge increase in traffic. This study presents a novel scheme to improve LoRaWAN scalability when multiple gateways are used without incurring excess backhaul traffic. The proposed scheme uses interflow network coding to encode traffic moving from gateways to network servers. The performance of the proposed scheme is evaluated via simulations. The results show that tremendous bandwidth savings can be achieved using the proposed scheme while allowing deploying more gateways to improve network scalability.
A comprehensive repair scheme for distributed storage systems
2023, Computer Networks
Modern data storage systems apply erasure codes to provide data reliability efficiently. Previous studies proposed a series of techniques to weigh repair/storage costs, reduce codec complexity, minimize repair time, improve fault tolerance, and enforce system-level service level agreement. These techniques have been designed in isolation, leading to performance limitations. We explore the potential advantages of combining these techniques to meet data storage systems’ requirements better and provide superior system performance. This work proposes a comprehensive repair scheme for fault data in distributed storage systems. First, we tailor design erasure codes in the presence of heterogeneity of storage devices. The core idea is to monitor device performance (e.g., access speed, reliability), compute two coefficients for each device, and use them to select the appropriate devices to create stripes of erasure codes. Second, we leverage the system hierarchy to perform intermediary repair operations, further minimizing cross-rack repair bandwidth. Finally, we propose a new repair scheme adapted to the skew of data access. To demonstrate the effectiveness of our comprehensive repair scheme, we evaluate various erasure codes via mathematical analysis and experiments in the Ceph cluster. In the mise-en-scène of traditional re-encoding methods and more recent adaptive erasure codes, our scheme stands out with significant savings in recovery bandwidth, code-switching bandwidth, repair time, and code-switching time.
Enhancing network lifespan in wireless sensor networks using deep learning based Graph Neural Network
2023, Physical Communication
Inadequate energy of sensors is one of the most significant challenges in the development of a reliable wireless sensor network (WSN) that can withstand the demands of growing WSN applications. Implementing a sleep-wake scheduling scheme while assigning data collection and sensing chores to a dominant group of awake sensors while all other nodes are in a sleep state seems to be a potential way for preserving the energy of these sensor nodes. When the starting energy of the nodes changes from one node to another, this issue becomes more difficult to solve. The notion of a dominant set-in graph has been used in a variety of situations. The search for the smallest dominant set in a big graph might be time-consuming. Specifically, we address two issues: first, identifying the smallest possible dominant set, and second, extending the network lifespan by saving the energy of the sensors. To overcome the first problem, we design and develop a deep learning-based Graph Neural Network (DL-GNN). The GNN training method and back-propagation approach were used to train a GNN consisting of three networks such as transition network, bias network, and output network, to determine the minimal dominant set in the created graph. As a second step, we proposed a hybrid fixed-variant search (HFVS) method that considers minimal dominant sets as input and improves overall network lifespan by swapping nodes of minimal dominating sets. We prepared simulated networks with various network configurations and modeled different WSNs as undirected graphs. To get better convergence, the different values of state vector dimensions of the input vectors are investigated. When the state vector dimension is 3 or 4, minimum dominant set is recognized with high accuracy. The paper also presents comparative analyses between the proposed HFVS algorithm and other existing algorithms for extending network lifespan and discusses the trade-offs that exist between them. Lifespan of wireless sensor network, which is based on the dominant set method, is greatly increased by the techniques we have proposed.
Optimal defense of a distributed data storage system against hackers’ attacks
2020, Reliability Engineering and System Safety
Citation Excerpt :
They have also become commercially attractive because of their large-scale data storage ability comparing with traditional data storage systems [4,5]. Distributed data storage systems have the potential to enhance data availability since multiple copies of a same data part can be made and allocated onto different computers to provide redundancy [6,7]. They are widely used in many fields such as financial transactions, military affairs and medical areas [8].
Distributed data storage systems are widely used to store data with the development of big data and cloud computing. Due to the complicated network environment, the data may risk being destroyed or stolen by an intentional hacker. In order to mitigate the risk of data destruction and data theft, this paper studies the joint optimization of data parts allocation and computers protections to defense a distributed data storage system in order to maximize the system reliability. The whole data is divided into multiple data parts, where copies of each part can be made and allocated onto different computers. Two different cases are considered, where all data parts on a computer will be destroyed in the first case and all data parts on a computer will not only be destroyed but also stolen in the second case, if the computer is intruded by the hacker. The system is reliable if the defender still has all data parts and the hacker does not obtain all data parts. For both cases, the system reliability is evaluated by extending the universal generating function technique. Numerical examples are carried out to illustrate the applications.
Fault tolerance in cloud computing environment: A systematic survey
2018, Computers in Industry
Citation Excerpt :
c) Expected Transmission Time Computation- It generates a matrix representing the communication cost between any two nodes in the network. ( d) k-out-of-n Allocation- Data is partitioned into n fragments using erasure code algorithm [78] and stored in the network such that retrieving k fragments would consume minimum energy. ( e) k-out-of-n Processing- It creates a job consisting of m tasks and schedules them on n processing nodes such that energy consumption would be minimized.
Fault tolerance is among the most imperative issues in cloud to deliver reliable services. It is difficult to implement due to dynamic service infrastructure, complex configurations and various interdependencies existing in cloud. Extensive research efforts are consistently being made to implement the fault tolerance in cloud. Implementation of a fault tolerance policy in cloud not only needs specific knowledge of its application domain, but a comprehensive analysis of the background and various prevalent techniques also. Some recent surveys try to assimilate the various fault tolerance architectures and approaches proposed for cloud environment but seem to be limited on some accounts. This paper gives a systematic and comprehensive elucidation of different fault types, their causes and various fault tolerance approaches used in cloud. The paper presents a broad survey of various fault tolerance frameworks in the context of their basic approaches, fault applicability, and other key features. A comparative analysis of the surveyed frameworks is also included in the paper. For the first time, on the basis of an analysis of various fault tolerance frameworks cited in the present paper as well as included in the recently published prime surveys, a quantified view on their applicability is presented. It is observed that primarily the checkpoint-restart and replication oriented fault tolerance techniques are used to target the crash faults in cloud.
Robust decentralized data storage and retrieval for wireless networks
2017, Computer Networks
In this paper we study the problem of distributed data storage using Rateless codes for large-scale resource-constrained wireless networks. We focus on building a robust storage system using Fountain codes from physically decentralized sources. Fountain codes, e.g. LT-codes, can achieve reduced complexity of both encoding and decoding, which caters well to the nature of such networks. We propose an energy-efficient distributed dissemination and coding scheme to build a decentralized LT-codes based storage over a network of resource-limited nodes to provide data survivability against possible failures. In the proposed scheme, each sensor node is assigned a selection probability derived from a Robust Soliton Distribution (RSD) in a distributed fashion. Source nodes then disseminate their data over the storage network randomly, in accordance with the selection probabilities. The proposed scheme is compared to similar schemes in the literature by means of simulations. We evaluate the energy required for building the storage as well as the energy needed for data retrieval from the storage system. Results show that energy consumption can be substantially reduced while achieving the required storage requirements.

View all citing articles on Scopus

Hossam S. Hassanein is a leading authority in the areas of broadband, wireless and mobile networks architecture, protocols, control and performance evaluation. His record spans more than 500 publications in journals, conferences and book chapters, in addition to numerous keynotes and plenary talks in flagship venues. He has received several recognitions and best papers awards at top international conferences. He is also the founder and director of the Telecommunications Research Lab (TRL) at Queen’s University School of Computing, with extensive international academic and industrial collaborations. He is a senior member of the IEEE, and is a former chair of the IEEE Communication Society Technical Committee on Ad hoc and Sensor Networks (TC AHSN). He is an IEEE Communications Society Distinguished Speaker (Distinguished Lecturer 2008–2010).

View full text

Distributed Data Storage Systems for Data Survivability in Wireless Sensor Networks using Decentralized Erasure Codes

Abstract

Introduction

Section snippets

Background and related work

Decentralized Erasure Codes for Data Survivability (DEC-DS)

Routing and energy efficiency

Performance evaluation

Conclusion

ACM Trans. Sen. Netw.

IEEE/ACM Trans. Netw.

Distributed mutual exclusion algorithms for intersection traffic control

IEEE Trans. Parallel Distrib. Syst.

Cloud computing applications for smart grid: A survey

IEEE Trans. Parallel Distrib. Syst.

Aggregated-proof based hierarchical authentication scheme for the Internet of Things

IEEE Trans. Parallel Distrib. Syst.

CCN-WSN - A lightweight, flexible content-centric networking protocol for wireless sensor networks

Proceedings of the 2013 IEEE Eighth International Conference on Intelligent Sensors, Sensor Networks and Information Processing

Ubiquitous access to distributed data in large-scale sensor networks through decentralized erasure codes

Proceedings of the Fourth International Symposium on Information Processing in Sensor Networks (IPSN ’05), Piscataway, NJ, USA

Distributed data storage in sensor networks using decentralized erasure codes

Proceedings of the Thirty-eighth Asilomar Conference on Signals, Systems, and Computers

Decentralized erasure codes for distributed networked storage

IEEE Trans. Inf. Theory

Network information flow

IEEE Trans. Inf. Theory

Network coding: An instant primer

ACM SIGCOMM Comput. Commun. Rev.

Linear network coding

IEEE Trans. Inf. Theory

Erasure coding vs. replication: A quantitative comparison

Proceedings of the First International Workshop on Peer-to-Peer Systems (IPTPS ’01)

How good is random linear coding based distributed networked storage

Proceedings of the First Workshop on Network Coding, Theory, and Applications (NetCod 2005)

An overview of algorithms for network survivability

ISRN Commun. Netw.

Enhancing survivability in virtualized data centers: A service-aware approach

IEEE J. Sel. Areas Commun.

A digital fountain approach to reliable distribution of bulk data

Proceedings of the ACM Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication (SIGCOMM ’98)

Polynomial codes over certain finite fields

J. Soc. Ind. Appl. Math.

LT codes

Proceedings of the Forty-third IEEE Annual Symposium on Foundations of Computer Science