Fessel, KarlDietrich, AndréZug, SebastianReussner, Ralf H.Koziolek, AnneHeinrich, Robert2021-01-272021-01-272021978-3-88579-701-2https://dl.gi.de/handle/20.500.12116/34705The key to IoT applications' success is the opportunity to exploit data generated by one node for various applications. Solutions for this are either centralized server systems, which aggregate the data and answer corresponding requests from different clients, or the concepts of edge computing, in which individual nodes take over the provision and processing of data directly. Although the advantages of immediate processing are obvious, edge computing concepts have so far been limited to more powerful nodes. Embedded in the DoRIoT project, we transfer the idea to low performance devices. This includes challenging questions related to security, scheduling and coordination issues. Additionally, we have to support the programming process itself. In order to achieve sufficient acceptance in the programming community we have to ensure that “freely programmable” is not bounded by hardware oriented programming paradigms and languages. Furthermore, the developer should be able to implement IoT-requests based on standard building blocks in a programming language of his choice. In this paper we introduce the architecture and a tool-chain to cope with these challenges based on a WebAssembly-interpreter (WAMR) embedded in the DoRIoT software stack. The prototypical integration provides the applicability of WASM compiler tool-chain, originally focused on web-applications, and supports the orchestration of multiple requests in parallel.enIoTRIOTDoRIoTWASM10.18420/inf2020_1161617-5468