SMEDE is TECoSA’s largest core project. To address the challanges of the emerging industrial applications in an edge computing environment, the SMEDE project applies novel approaches to managing communication and computing resources. Ensuring security in the highly distributed edge computing environment, and providing sufficient isolation in multi-tenant environments, are two of the projects’s main focus areas.
The digitalization of industries is transforming our society, including development processes as well as products and services. Performed in the right way, industrial digitalization leads to lower production costs, higher efficiencies as well as the realization of new use cases with significant value-adds. Its benefits increase with the amount of digitized objects and processes, which dramatically increase the amount of involved software and data volumes, raising the crucial question where the data will be processed and stored.
Edge computing could potentially address the needs of industrial digitalization, as it would enable to have small amount of resources in the devices, but at the same time it overcomes the latency and bandwidth issues related to centralized cloud-based solutions. Due to these inherent properties, edge computing systems have a tremendous potential.
It is, however, apparent that edge computing can only serve as an enabler of industrial digitalization if it can meet the application requirements in terms of latency, throughput and security, while being cost-efficient.
Challenges and approaches
Meeting the requirements of emerging industrial applications in an edge computing environment presents a number of challenges. Providing latency and throughput guarantees comparable to those in embedded systems requires the novel approaches to managing communication and computing resources, as existing solutions focus on either one or the other. Guaranteeing latency and throughput in multi-tenant environments also requires that the resources be shared among different users, with different workload characteristics, in an efficient way. Cloud-based approaches to abstracting computational resources, e.g., virtual machines, introduce significant overhead, while container-based abstractions may not provide enough performance isolation. These problems are tightly coupled to the TECoSA focus area Predictability.
Ensuring security in an edge computing environment is challenging due to the highly distributed nature of edge environments, which results in an extended attack surface compared to cloud environments. In addition, lightweight virtualization that enables efficient resource management may not provide sufficient isolation in a multi-tenant environment. These problems are tight coupled to the TECoSA focus area Security.
Addressing these challenges requires new architectures and abstractions to be developed, together with corresponding methodologies, which allow the analysis and optimization of the edge infrastructure, to achieve a self and context-aware smart edge infrastructure. Beyond methodologies and algorithms, experimental research is essential for evaluating and validating novel solutions in controlled environments, but at the same time with realistic workload