CPS Orchestrator with Distributed Image Processing Pipeline
Contributors
Researcher (4):
Description
This repository contains a CPS workload orchestrator along with a distributed image processing pipeline developed as part of the
Heterogeneous Application Orchestration in Cyber-Physical Systems (DAIS 2026) paper submission. The system demonstrates NFR-driven adaptive
orchestration across heterogeneous devices and execution formats in a reproducible way. It is therefore provided as open research output in the
form of a software artefact.
CPS offer heterogeneous compute resources in the form of a computing continuum with physical constraints. These constraints extend beyond
typical compute characteristics (processors, memory) towards power, heat and other external factors, and must be reflected by appropriate
software architectures and a translation of these architectural decisions into runtime adaptation decisions. Exploiting them in the appropriate
manner leads to sustainable software operation, one of the goals of the TaSSAreCt research project (https://tassarect.github.io/) – Towards
Sustainable Software Architectures for Cyber-physical Systems of Systems.
The details of the orchestrator and the exemplary pipeline as well as steps to undertake in reproducibility and repeatability studies are described
in the file PIPELINE.md
Files
cps-orchestrator.zip
Files
(287.3 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:4faa587924671cecc1bda2dcd58f54d3
|
287.3 kB | Preview Download |
Additional details
Funding
- Swiss National Science Foundation
- Towards Sustainable Software Architectures for Cyber-physical Systems of Systems (TaSSAreCt) 10002384
Software
- Programming language
- Python