Driving Industry 4.0 at Distributed Edges with Cloud Orchestration
multiple production plants in different locations . Each plant needs to run an edge infrastructure within its own premise . Applications need to be deployed into each plant turning it into a distributed system . More granular scenarios can be achieved by deploying more than one edge infrastructure into a plant . For example , by deploying an edge infrastructure into each production line and all work centers of a single plant [ 18 ]. An edge infrastructure building an edge node consists of :
- hardware needed to provide the necessary computing resources ( local processing and local storage ),
- software components such as operating system , hypervisor , container runtime , and containerized business applications and backing services , and
- interfaces supporting industry standards ( e . g ., MQTT , OPC-UA , TCP / IP , REST ) that establish bi-directional communication and interactions between cloud-edge layer , between edge nodes within the edge layer ( east / west communication ) as well as between assetsedge layer .
Fig . 3 - Approach for ensuring business continuity for mission critical applications based on [ 14 ].
As both cloud and edge provide cloud native capabilities such as support to container runtimes and containerized applications , the same business applications can run on the cloud or on the edge nodes in a distributed manner . This empowers solution providers with the flexibility to determine the optimal balance of cloud and edge functionality needed to fulfill their scenarios . The deployment of the containerized applications can vary by customer , depending on their infrastructure ( e . g ., edge hardware computing capabilities and footprint constraints ) and needs . This
IIC Journal of Innovation - 25 -