Intelligent urban planning for future smart cities
In the digital transformation, edge computing plays a crucial role in the further development of technologies and services. The near real-time processing of data directly at the edge of the network enables innovative applications and significantly improves the performance of systems. A significant contribution to this field of research was made by the C4AI Edge project, a collaborative initiative between fortiss and IBM in 2022 as part of the joint C4AI research center.
The project focused on the exploration of context-aware Edge-Cloud service and data orchestration, combining the strength of IBM Pathfinder with the fortiss expertise on decentralised and context-aware Edge computing, to develop context-aware Industrial Internet of Things (IIoT) services.
IIoT applications are today deployed across the so-called Edge-Cloud continuum, and the overall management of the applications during setup and runtime is handled by orchestration tools such as Kubernetes. These orchestrators, originally devised to support Cloud-based deployment, fall short in the required learning and adaptation capabilities and are unable to adjust and track the overall data workflow.
In this six-month exploratory project, fortiss and IBM explored the use of context-awareness (fortiss) and intelligent data discovery and linkage (IBM Pathfinder) to achieve automated orchestration of Edge-Cloud IIoT services for a manufacturing use-case.
The concept developed by fortiss is available in an open-source demonstrator Movek8s.
fortiss and IBM explored together the pathways to achieve a new type of orchestration that can integrate the current purpose (reduce latency) and also comply with data protection regulations.
To achieve these goals, partners have explored the type of metadata that may be considered as “context”, focusing on a Manufacturing use-case, “IoT shop-floor to Cloud Data flow”, which embodies a usual manufacturing scenario. It considers heterogeneous IoT data sources (environmental sensors, cameras), and considers a single cluster deployment, involving two different Edge nodes and a Cloud node.
The application workload has been set based on the fortiss TSMatch open-source middleware, developed to provide environmental sensing on the Edge. A simple embodiment of context-awareness has been set to capture external aspects of the system, namely, worker location, and explored the move of the application workload to the location closer to the end-user, to reduce latency. A proof-of-concept demonstrator is available to the research community on the fortiss IIoT Lab (fortiss Movek8s demonstrator).
The proposed concept sets the initial pathway for a new type of IIoT application management, that takes into consideration application requirements such as bounded latency, Central Processing Unit (CPU), or memory but also considers advanced aspects related with the data worklow.
The work developed continues as one of the key innovation blocks of the Horizon Europe Cognitive Decentralised Edge Cloud Orchestration (CODECO, 2023-2025), coordinated by fortiss and involving IBM.
The project was carried out in collaboration with IBM Research Zurich Hybrid Cloud.
