Paper accepted at ICSE 2025
2025/02/03

Why is the dynamic adaptation of microservice chains important? In mission-critical applications such as autonomous multi-task perception, healthcare, or robotic search and rescue operations, systems must be able to respond quickly and efficiently to changes. The challenge lies in dynamically adjusting such complex microservice chains during runtime, considering multiple requirements.
In this context, we are pleased to announce that researchers from TK, in collaboration with Huawei, will present their paper titled “MARQ: Engineering Mission-Critical AI-based Software with Automated Result Quality Adaptation” at the renowned International Conference on Software Engineering (ICSE 2025). This project investigates how microservice chains can be designed to meet the demands of mission-critical applications, with a focus on dynamic runtime adaptation, especially in AI-based systems that require real-time decisions.
The research findings impressively demonstrate how decision-making can be significantly accelerated, opening up the possibility of adjusting microservice chains in real-time. This capability is essential for numerous applications, whether in autonomous perception, critical medical systems, or the execution of complex rescue missions. Through MARQ, the foundation is laid for more powerful, flexible, and adaptable software that is up to the challenges of these scenarios.
We warmly congratulate the researchers on this success and look forward to further developments in this forward-looking project.
Citation information:
[1] Uwe Gropengießer, Elias Dietz, Florian Brandherm, Achref Doula, Osama Abboud, Xun Xiao, and Max Mühlhäuser. 2025. MARQ: Engineering Mission-Critical AI-based Software with Automated Result Quality Adaptation [to appear].
