Performability of Network Service Chains: Stochastic Modeling and Assessment of Softwarized IP Multimedia Subsystem

Service provisioning mechanisms implemented across 5G infrastructures take broadly into use the network service chain concept. Typically, it is coupled with Network Function Virtualization (NFV) paradigm, and consists in defining a pre-determined path traversed by set of softwarized nodes to provide specific services. A well known chain-like framework IP Multimedia Subsystem (IMS), key infrastructure networks, that we characterize both performance an availability perspective. Precisely, supported designed from scratch testbed realized through Clearwater platform, perform stochastic assessment IMS (softIMS) architecture where two main stages stand out: i) analysis, where, exploiting queueing decomposition method, formalize optimization problem resource allocation modeling each softIMS node as $M/G/c$ system; ii) assessment, adopting Stochastic Reward Net methodology, are able behavior terms failure/repair events, derive optimal configurations satisfying given requirement (e.g., five nines) while minimizing deployment costs. Two routines dubbed OptCNT OptSearchChain have been devised govern analyses, respectively.