Driving Industry 4.0 at Distributed Edges with Cloud Orchestration Potentials and Market Chances for Edge Computing Solutions
Edge computing is attracting a huge market potential due to its promising business value proposition . According to IDC Research [ 4 ] , by 2023 “ 70 % of IoT deployment will include edge-based decision making , supporting organizations ' operational and strategic agendas and 70 % of enterprises will run varying levels of data processing at the IoT edge . In tandem , organizations will spend over $ 16 billion on IoT edge infrastructure in that time ”. This estimation is not a surprise since the number of connected assets along the D2O phases growing .
A smart factory will be producing on average “ 5 petabytes of data per day ” [ 5 ]. With 5G cellular networks , a higher volume of data produced by machines and robots can be available with higher throughput for real time scenarios in manufacturing such as smart sensing . In this context , such scenarios are highly dependent on “ human-like latency and always-on connectivity ” [ 6 ]. Analogically , the same capabilities are mandatory in autonomous transportation ( e . g ., autonomous driving ). Both manufacturing and autonomous transportation scenarios will profit from artificial intelligence ( AI ) at large , and more specifically from machine learning ( ML ) and deep learning ( DL ) applications using edge computing capabilities to enable instantaneous decision making and optimize the processes in real time .
Furthermore , use cases and deployment scenarios that require real-time decision-making support within verticals ( healthcare , smart cities , autonomous vehicles , smart retail ) are augmented and virtual reality ( AR / VR ), Visual Inspection , Cloud Gaming [ 7 ]. In context of Industry 4.0 , edge computing supports the realization of Intelligent Assets , Intelligent Factories , Intelligent Logistics , Intelligent Products , and Empowered People scenarios [ 3 ] into a holistic , intelligent supply chain . Further introduction to and description of Edge Computing for industrial applications can be found in the literature [ 8 ] [ 9 ] [ 10 , 11 ].
Challenges of Edge Computing
Edge computing offers compelling benefits for Industry 4.0 scenarios . It also presents challenges that must be addressed .
From Few to Many Distributed Edge Nodes – Higher Complexity . Edge computing is characterized by highly distributed topologies that often consist , depending on the scenario , of edge nodes with limited computing and storage . At the same time , edge landscapes are highly heterogeneous in terms of platforms , infrastructure , hardware networking , and security [ 12 ]. Both characteristics increase the complexity of deployment , management , and orchestration of edge solutions . Heterogeneous landscapes raise some challenges such as interoperability and plug and play solutions to ensure the continuity of business-critical processes . To maximize business value , edge computing deployments need to be simplified to ensure that the edge focuses only on the business-critical and edge-specific components of business applications being extended from the cloud .
IIC Journal of Innovation - 21 -