Core slicing: dedicated vs shared NFs

Three Models for Core Slicing

Core slicing determines how 5G Core network functions are deployed across slices, and three distinct models exist. The fully shared model runs all NFs in one pool with slices differentiated only by S-NSSAI tagging in messages. This is cheap but provides no real isolation between slices. The fully dedicated model gives each slice its own complete set of NFs, achieving maximum isolation but at three times the infrastructure cost for three slices. The hybrid model, which is the standard 5G approach, shares common NFs while dedicating session-handling NFs per slice.