Model for Jointly Determining Service Function Chain Routes and Update Scheduling with State Consistency

Abstract

Service chaining technology realizes that packets are processed through virtual network functions (VNFs); the set of successive VNFs is called a service function chain (SFC). Routes of SFCs need to be updated in order to resolve a quality of service degradation due to concentration of processing loads on certain VNFs. When updating the SFC routes, states of migrated VNF instances need to be kept consistent to prevent a degradation of processing accuracy of the VNF instances. Existing studies determine SFCs to be updated, their routes, and an update scheduling in multiple phases; each decision does not necessarily minimize the total time required to update the SFC routes. This paper proposes a model that jointly determines SFCs to be updated, their routes, and an update scheduling while guaranteeing state consistency. The proposed model considers the time required to update flow entries of SFC routes, the time required to migrate states of VNF instances, and the delay time required to transmit, propagate, and process packets. The objective function is to minimize the total time required to update SFC routes under a constraint of processing load balancing among VNF instances. The proposed model is formulated as an integer linear programming problem. The proposed model is compared to a benchmark model based on the existing studies. Numerical results show that the proposed model can reduce the total update time compared to the benchmark model.