Work

Resilience in the Internet of Things for Smart City Applications

Year: 2021
Type: dissertation
Abstract: Nowadays companies and governments are using Information and Communication Technologies as tools to deploy their services and make them accessible to citizens; in order to expand urban resource efficiency with a low environmental impact and contributing with the development of the economy. This trending is known as the Smart City paradigm and it has taken advantage of the Cloud to Internet of Things continuum to provide communication and collect a massive amount of data. This scenario permits that the data can be processed and analyzed by applications to support smart services, enabling heavy calculations that run inside powerful data centers in the Cloud. The Cloud to Internet of Things not only offers power to process data, but also makes it possible to use virtualization technologies to enhance how the object heterogeneity could be managed. Services like eHealth, smart traffic control, and smart home applications are composed of different functions that can be virtualized over physical hardware components within the network landscape. These Virtual Functions are grouped in a structure called Service Chains that fulfill particular smart service requirements, enabling a new broader set of smart end-user applications. The Cloud to Internet of Things continuum infrastructure provides communication, processing, and storage support for these applications. However, this complex, heterogeneous, and distributed landscape requires orchestration and management mechanisms in order to guarantee their proper functioning, especially in the face of failures. One particular factor to manage is the resilience to provide service availability even in the event of failures. Automated proactive solutions to enhance the survivability of Service Chains when failures occur have to be considered within the orchestration solution for the Cloud to Internet of Things continuum. This research proposes an architecture and a set of mechanisms to orchestrate, formalize, and embed a collection of service requests for chaining Virtual Functions jointly to fulfill specific requirements of applications while enhancing their resilience. The first contribution of this work is a resilience architecture for the deployment of Internet of Things services and applications in Smart Cities. The architecture proposed takes advantage of virtualization techniques to deal with the heterogeneity in the Cloud to Internet of Things continuum, as well as to provide a set of components focused on improving the resilience of applications. Specifically, the Resilience Manager module in the architecture implements a set of functions to enhance the availability of applications in case of failures. The architecture proposed worked as an inspiration for the design and implementation of an ontology to describe the Internet of Things infrastructure, to standardize how the information of the underlying components is exchanged. The second set of contributions of this work are focused on the Resilience Manager module of the proposed architecture. In detail, a framework to address the composition and embedding of Service Chains in the Cloud to Internet of Things continuum with support of replicas to increase the availability of applications was designed and implemented. The composition element in the framework is tackled via a formal grammar that enables the description of customized applications, allowing the definition of replicas for their components. In a second step, once the applications were specified, a Pareto analysis is used to optimize the selection of the components of the applications according to a given set of goals. Regarding the embedding of Virtual Functions in the substrate infrastructure, three contributions are proposed: (1) an Integer Linear Programming model that prioritizes the use of nodes with higher availability; (2) a genetic algorithm that uses a fitness function that combines node availability, use of disjoint nodes, and the tiers to which the nodes belong to; and (3) a heuristic to handle more complex scenarios by taking advantage of the multi-tier scenario comprising the Cloud-Fog-Mist-Internet of Things. The embedding mechanisms proposed in this work were evaluated via simulation. The assessment included measurements of failure rate, node utilization, and response time of the Service Chains embedded in the subtracted infrastructure. Simulation results show that it is possible to increase the resilience of chained Virtual Functions, while balancing the load of the infrastructure nodes.
Author: Abreu, David Alejandro Perez
Language: English
Open Access status: green
OpenAlex ID: W7024268534
Cites: 103
Cited by: 1
Related to: 10