Introduction Research Background
Since 2020, academia and industry have paid much more efforts to investigate 6G networks. A large number of white papers and reports have been released in order to push 6G research ahead. Com- paring with previous five generations of networks, 6G networks are designed to provide three-dimensional (3D), full-coverage, seamless and tailored services [1]. That is to say, 6G networks are not limited in the traditional terrestrial region. Satellite and aerial regions [2] are incorporated as well (Fig. 1). Consequently, millions of dedicated network appliances (e.g., servers, base stations, drones, flying airships, satellites) are accepted as fundamental components of 6G networks. Due to the huge amount of appliances, the topology of 6G networks will be complicated. If researchers and industrial enterprises insist on adopting the traditional networking method of deploying new dedicated appliance, it is hard to manage, maintain and develop new emerging services. Based on these backgrounds, the network softwarization (NetSoft) [2], [3], mainly achieved by SDN, NFV and network slicing (NS) technologies, is widely accepted as the dominant attribute of 6G networks. With NetSoft, network appliances will be softwarized and decoupled into software blocks and general-purpose hardware. Software blocks indicate network function (NF, e.g., firewall) blocks and resource (e.g., computing, storage) blocks. Users can initiate tailored service requests and their rigid performance metrics to their contracted telecommunication providers (TSPs). TSPs are responsible for realizing their tailored services and fulfilling rigid performance metrics. TSPs can chain these software blocks and constitute the tailored service requests on top of 6G general-purpose hardware. Apparently, this serving scheme in 6G networks will bring about huge benefits and has great potential, in terms of service upgradation and deriving extra new tailored services. Much research attention has been found in the literature, especially those related to softwarized resource allocation of tailored services [3].