Abstract
Modeling a complete Internet of Things (IoT) infrastructure is crucial to assess its availability and security characteristics. However, modern IoT infrastructures often consist of a complex and heterogeneous architecture and thus taking into account both architecture and operative details of the IoT infrastructure in a monolithic model is a challenge for system practitioners and developers. In that regard, we propose a hierarchical modeling framework for the availability and security quantification of IoT infrastructures in this paper. The modeling methodology is based on a hierarchical model of three levels including (i) reliability block diagram (RBD) at the top level to capture the overall architecture of the IoT infrastructure, (ii) fault tree (FT) at the middle level to elaborate system architectures of the member systems in the IoT infrastructure, and (iii) continuous time Markov chain (CTMC) at the bottom level to capture detailed operative states and transitions of the bottom subsystems in the IoT infrastructure. We consider a specific case-study of IoT smart factory infrastructure to demonstrate the feasibility of the modeling framework. The IoT smart factory infrastructure is composed of integrated cloud, fog, and edge computing paradigms. A complete hierarchical model of RBD, FT, and CTMC is developed. A variety of availability and security measures are computed and analyzed. The investigation of the case-study’s analysis results shows that more frequent failures in cloud cause more severe decreases of overall availability, while faster recovery of edge enhances the availability of the IoT smart factory infrastructure. On the other hand, the analysis results of the case-study also reveal that cloud servers’ virtual machine monitor (VMM) and virtual machine (VM), and fog server’s operating system (OS) are the most vulnerable components to cyber-security attack intensity. The proposed modeling and analysis framework coupled with further investigation on the analysis results in this study help develop and operate the IoT infrastructure in order to gain the highest values of availability and security measures and to provide development guidelines in decision-making processes in practice.
Highlights
Internet of Things (IoT) has recently emerged as a mainstream computing infrastructure with a rapidly growing interest both in industry and academia due to the pervasiveness and ubiquitous features of IoT sensors/devices which play the role of a software-defined gateway to the physical world [1]
Stringent requirements in design and implementation of most IoT infrastructures are that (i) applications and services running on the infrastructure are mostly latency-sensitive, (ii) massive data generated by IoT sensors/devices is seamlessly synchronized, stored, and processed across the infrastructure often at a huge data volume, a high data transaction rate and with an uncertain data variation, and (iii) operations of the IoT infrastructure at all levels must strictly satisfy quality of service (QoS) terms in service level agreement (SLA) including data accuracy, availability and security, operational costs, and user expectations
State-of-the-art computing paradigms: Cloud computing has been accredited as a centralized computing paradigm successfully adopted in many online business services and applications in the past few years featured by its pricing models of pay-as-you-go and its service platforms of Software as a Service (SaaS), Platform as a Service (PaaS) and Infrastructure as a Service (IaaS) [24,25]
Summary
Internet of Things (IoT) has recently emerged as a mainstream computing infrastructure with a rapidly growing interest both in industry and academia due to the pervasiveness and ubiquitous features of IoT sensors/devices which play the role of a software-defined gateway to the physical world [1]. An IoT infrastructure often features with a huge number of heterogeneous Internet-connected objects (e.g., IoT sensors/actuators) at the edge while the infrastructure’s computing backbones are powerful computing paradigms (e.g., cloud/fog/edge) that come along with specific functionalities such as data analysis and/or recommendation [3]. Afterwards, existing system architectures in previous works for integration and interoperability of those computing paradigms in different practical applications are presented to realize the currently existing gap of studies in literature on exploring the integration and interoperability of cloud/fog/edge computing paradigms at once in an IoT infrastructure. Featured by the nature of decentralized and open computing, the fog computing paradigm opens opportunities to disperse computing and storage power to the edge of the computing network for notable improvements of latency sensitivity, data processing performance, connectivity, scalability, and adaptability of services and applications in local areas [30]. The concept of fog computing is roughly considered as a combination of Mobile Cloud Computing (MCC) and Mobile
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
