Performability Evaluation of Softwarized Multi-Tenant 5G Architectures with Rejuvenation

In this work, we devise a framework designed to evaluate the performability of softwarized multi-Tenant 5G architectures, where rejuvenation might be applied. Rejuvenation refers to the procedure of occasionally restarting a software system aimed at counteracting aging-related problems and improving its availability. We utilize queueing theory for a performance characterization, employing an M / G / k model to embody estimated service times of 5G nodes. Conversely, availability modeling is addressed through the Multidimensional Universal Generating Function (MUGF), a recently developed technique particularly beneficial for multi-Tenant architectures with numerous sub-elements. Most numerical parameters have been estimated through Open5GS, an open-source platform emulating the softwarized core network of a real 5G network operator.