Robust service provisioning of service function chain under demand uncertainty

In network function virtualization, resource allocation is one of the key techniques to ensure QoS service requests in face uncertain traffic demand and fluctuation user services. The previous related works usually assume that a deterministic value then reallocate substrate resources deal with uncertainty during operation. response performance degradation caused by fluctuation, paper models chain orchestration under as robust optimization problem, where parameter Γ introduced control conservativeness solution. On this basis, strong duality theory used transform original problem into mixed-integer linear programming, devise an exact Robust Service Provisioning (RSP) algorithm. simulation evaluation demonstrates proposed algorithm could achieve different levels robustness make trade-off between cost. impact on realized price also analysed. Thus, get effective scheme demands provide reference total