Edge to Cloud Network Function Offloading in the ADAPTO Framework

As telcos increasingly adopt cloud-native solutions, classic resource management problems within cloud environments have surfaced. While considerable attention has been directed toward the conventional challenges of dynamically scaling resources to adapt to variable workloads, the 5G promises of Ultra-Reliable Low Latency Communication (URLLC) remain far from being realized. To address this challenge, the current trend leans toward relocating network functions closer to the edge, following the paradigm of Mobile Edge Computing (MEC), or exploring hybrid approaches. The adoption of a hybrid cloud architecture emerges as a solution to alleviate the problem of the lack of resources at the edge by offloading network functions and workload from the Edge Cloud (EC) to the Central Cloud (CC) when edge resources reach their capacity limits. This paper focuses on the dynamic task offloading of network functions from ECs to CCs within cloud architectures in the ADAPTO framework.