A vision on Smart, Decentralised Edge Computing Research Directions

Zusammenfassung

Organisations across a variety of industries rely today on the Internet of Things (IoT) to, on the one hand, increase operational efficiency and, on the other hand, to improve the overall decision making processes and increase the value-add of their products/business. From a computer science perspective, IoT is a system of interconnected cyber-physical devices, software, data, and people. The aim is to allow data to be accessible anywhere and anytime, so IoT relies on Cloud computing to provide this support from a storage perspective. The data accessibility and computational power that the Cloud provides has led to an increase in the generated data, which had, as consequence, an increase in energy consumption, among other aspects. Cloud-based services are highly relevant in case an organisation aims at extending storage capacity, or to allow for remote work operations. These are aspects that IoT systems require, but in addition to data storage, the key aspects in IoT are data processing and analysis. The data processing and analysis are therefore provided via the Cloud, in a centralised way. IoT, being based on the Internet, requires support for service and data space decentralisation, to further evolve. Predictions point out to having 75% of enterprise data supported by Edge computing in 20251. Relevant to this evolution is to discuss the different perspectives of Edge computing and Edge-Cloud; and to have a perspective on the most recent developments in Europe concerning Next Generation IoT and Edge computing. This white paper is therefore a tool expected to assist in such cross-exploration, via the clarification of Edge computing notions, a brief introduction to NGIoT flagship projects, their current standardisation and pre-standardisation efforts. For this purpose, this white paper provides the following contributions.: • To provide an overview on different Edge computing concepts, contributing to a better understanding of concepts such as “far Edge”, or “near Edge”. • To explain the EU-IoT Edge computing functionality in terms of the 4 identified EU-IoT scope areas: Human/IoT interfaces; far Edge; near Edge; Infrastructure; Data Spaces. • To discuss existing Edge research directions in active NGIoT European projects that are under the umbrella of EU-IoT. • To provide an initial perspective on current Edge standardisation efforts being taken by the projects under the umbrella of EU-IoT.