Programmability and softwarization advocate the emerging era of open-source platforms, which embraced by both industry and academia is foreseen as a vital pillar in the construction of next generation mobile networks. Such a valuable open-source project is OpenAirInterface (OAI), which provides a standard compliant mobile network infrastructure, merely based on general purpose hardware computers. While OAI is nowadays widely used by industry and research institutes in proof-of-concept or commercial wireless testbeds, an analysis of the complex functions within the platform is yet to be performed in a large scale. We believe that further research is required to demystify the capabilities of existing tools and present guidelines that alleviate the enhancement and development of additional features. In this context, in this work we shed light on one of the crucial components of any mobile system, namely resource scheduling, while providing an analysis of the available code and instructions to ease the development of new scheduling algorithms based on OAI. Moreover, we demonstrate a performance evaluation of up to 10 UEs for existing and newly implemented scheduling algorithms. Results show, that the development of additional algorithms in OAI is achievable and the experimental behavior follows the theory. Our implementation and observations can serve as a basis for research in the field, and foster the elaboration of theoretical concepts and emerging 5G solutions in practical testbeds.
This demo presents a self-operating Kubernetes (K8s) cluster that uses digital twinning and machine learning to autonomously adapt its Horizontal Pod Autoscaler (HPA) to workload changes. The demo uses a digital twin of a K8s cluster to gather performance statistics and learn a model for the workload. With the model, the cluster autonomously adjusts HPA parameters for better performance. The demo illustrates this process and shows that the requested pod seconds decrease by ∼37 %, while the request latency stays mostly unaffected.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.