Algorithms and Techniques for Scalable, Reliable Edge-to-Cloud Industrial Internet of Things

Abstract

The Industrial Internet of Things (IIoT), which is a special class of Internet of Things (IoT), operates in large, distributed and dynamic environments comprising sensors all the way to large server clusters. IIoT is envisioned to support mission critical applications deployed in domains such as transportation, healthcare, manufacturing, and energy. Realizing the vision of IIoT requires scientific advances in the systems software for (a) the discovery and data dissemination between machines at the edge and the cloud, and (b) timely and reliable analytics conducted in the cloud for proactive maintenance and safety of the industrial systems that use IIoT. To address these requirements, this dissertation makes three contributions. First, it presents algorithms for a scalable discovery protocol as well as a coordination service in wide area network (WAN) environments. These algorithms are evaluated in the context of a standardized data-centric publish/subscribe messaging service called Object Management Group (OMG)’s Data Distribution Service (DDS). Second, it provides algorithms and a systems framework for highly available and real-time cloud infrastructures to satisfy the timeliness and reliability requirements of cloud-based data analytics. Finally, it provides a model-based testing automation framework for validating the performance of OMG DDS applications that must meet specific service levels through the use of different combinations of DDS quality of service (QoS) configurations.