Abstract:The emerging network-softwarization technologies, such as software-defined networking and network function virtualization play important roles in 5G communication and future networks. One of the critical challenges of the practical application of the softwarized networks is to appropriately place virtual network functions (VNFs). The underlying resources and traffic requirements are often factored in the previous works of VNF placement. However, VNFs’ dynamic abilities of changing traffic are usually ignored. Resources allocated to VNFs can vary their traffic-change ratios, and the adjustment of resource volumes should be provisioned to support traffic changes. In this work, we pay attention to the joint optimization problem of VNF Placement, CPU Allocation, and flow Routing (VNFPAR) in the scenarios consisting of VNFs that can dynamically change traffic. We employ the logarithmic functions to approximate VNFs’ traffic change relations and formulate VNFPAR as a mixed-integer nonlinear programming (MINLP) problem. We demonstrate that this problem is highly nonconvex and involves highly coupled variables. For small-scale VNFPAR problems, we propose an optimal algorithm based on relaxation and programming to consume the minimum bandwidth resources. Because VNFPAR is NP-hard, to quickly find near-optimal solutions for large-scale VNFPAR problems, we present heuristic algorithms based on multistage greedy and simulated annealing, respectively. Besides, to achieve a tradeoff between solution quality and execution time, we decompose VNFPAR into subproblems and design an alternating optimization-based method. We evaluate our algorithms on real-work topologies and traffic patterns. Extensive simulations show that our proposed heuristic algorithms are convergent, stable, and effective in terms of solving VNFPARs. The proposed algorithms have small optimality gaps within 7.4%–26.3%. Meanwhile, they save 39.3%–48.4% bandwidth resources compared with relevant baseline technologies.

On Orchestrating Virtual Network Functions in NFV

NFV-Inspector: A Systematic Approach to Profile and Analyze Virtual Network Functions

Service Function Chain Placement for Joint Cost and Latency Optimization.

Virtualized network functions resource allocation in network functions virtualization using mathematical programming

Cost-Efficient Virtual Network Function Graph (vnfg) Provisioning in Multidomain Elastic Optical Networks

Network Function Consolidation in Service Function Chaining Orchestration

Automated Function Placement and Online Optimization of Network Functions Virtualization

Robust Online Composition, Routing and NF Placement for NFV-enabled Services

Placement Optimization of Virtual Network Functions in a Cloud Computing Environment

Virtual network function placement and resource optimization in NFV and edge computing enabled networks.

Vnf Deployment And Flow Scheduling In Geo-Distributed Data Centers

Virtual Network Function Placement for Service Function Chaining with Minimum Energy Consumption

Network Function Virtualization in Elastic Optical Networks

Online Load Balancing for Network Functions Virtualization

Cost Efficient State-Aware Function Placement and Flow Scheduling for NFV Networks.

Optimizing NFV Chain Deployment in Software-Defined Cellular Core

Joint VNF Placement, CPU Allocation, and Flow Routing for Traffic Changes

An Approach for Service Function Chain Reconfiguration in Network Function Virtualization Architectures.

Virtual Network Function Placement with Function Decomposition for Virtual Network Slice

Virtual Network Function Placement Considering Resource Optimization and SFC Requests in Cloud Datacenter

Distributed Placement And Online Optimization Of Virtual Machines For Network Service Chains