Virtual Network Function Placement Model for Service Chaining to Relax Visit Order and Routing Constraints
Naoki Hyodo, Takehiro Sato, Ryoichi Shinkuma, Eiji Oki
Abstract
Naoki Hyodo, Takehiro Sato, Ryoichi Shinkuma, Eiji Oki
Abstract
Service chaining creates a desired service by the sequential concatenation of Virtual Network Functions (VNF). Relaxing the visit order constraints of requested VNFs and allowing the creation of loops in service chaining paths enable a reduction in the number of VNFs placed on the network. However, the problem of applying both techniques at the same time has never been studied. This paper proposes a VNF placement model that minimizes the costs associated with VNF placement and link utilization while supporting partially-ordered VNFs and service chaining paths with loops. We formulate the VNF placement model as an integer linear programming problem. Simulations quantitatively show that the proposed model reduces the cost compared to the conventional model.
OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Service chaining creates a desired service by the sequential concatenation of Virtual Network Functions (VNF). Relaxing the visit order constraints of requested VNFs and allowing the creation of loops in service chaining paths enable a reduction in the number of VNFs placed on the network. However, the problem of applying both techniques at the same time has never been studied. This paper proposes a VNF placement model that minimizes the costs associated with VNF placement and link utilization while supporting partially-ordered VNFs and service chaining paths with loops. We formulate the VNF placement model as an integer linear programming problem. Simulations quantitatively show that the proposed model reduces the cost compared to the conventional model.
Key concepts: Chaining, Virtual network, Computer science, Integer programming, Service (business), Distributed computing, Concatenation (mathematics), Routing (electronic design automation)