NFV Network Research Articles

Revealing the Impact of Dynamic Traffic on NFV Networks Planned Under Static Assumptions

Revealing the Impact of Dynamic Traffic on NFV Networks Planned Under Static Assumptions

Multi-provider NFV network service delegation via average reward reinforcement learning

In multi-provider 5G/6G networks, service delegation enables administrative domains to federate in provisioning NFV network services. Admission control in selecting the appropriate domain for service deployment, without prior knowledge of service requests’ statistical distributions, is fundamental to maximize average profit. This paper analyzes a general federation contract model for service delegation in various ways. First, under the assumption of known system dynamics, we obtain the theoretically optimal performance bound by formulating the admission control problem as an infinite-horizon Markov decision process (MDP) and solving it through dynamic programming, which is used as a benchmark to evaluate practical solutions. Second, we apply Reinforcement Learning (RL) to practically tackle the problem when the arrival and departure rates are not known. For the first time in this context, we analyze the performance of the widely used Q-Learning algorithm, and prove as it maximizes the discounted rewards, it is not an efficient solution due to its sensitivity to the discount factor. Then, we propose the average reward reinforcement learning approach (named “R-Learning”) to find the policy that directly maximizes the average profit. Finally, we evaluate different solutions through extensive simulations and experimentally using the 5Growth management and orchestration platform. Results confirm that the proposed R-Learning solution always outperforms Q-Learning and the greedy policies. Furthermore, while there is at most a 9% optimality gap in the ideal simulation environment, it competes with the MDP solution in the experimental assessment.

Application of an Innovative Convolutional/LSTM Neural Network for Computing Resource Allocation in NFV Network Architectures

An innovative resource allocation framework for virtualized network environments and based on the application of Artificial Intelligence techniques is proposed and investigated. It integrates the needed processing capacity prediction procedure and the allocation one which determines the capacity under-allocation or over-allocation needed to minimize the resource allocation and Quality of Service degradation costs. The proposed solution is based on: i) a monitoring procedure in which the processing capacities required by virtual instances are periodically monitored; ii) an integrated allocation/prediction procedure in which the processing capacities to be allocated to the virtual instances are evaluated in time intervals successive to the monitoring periods. This second procedure uses a Convolutional/Long Short Term Memory neural network whose loss function is defined so as to minimize an overall cost dependent on both the cloud resource allocation and Quality of Service degradation costs. We evaluate the proposed solution in backbone and metropolitan traffic and network scenario. We show how in the traffic scenario of an Italian Mobile Operator in Milan zone, the proposed solution far outperforms the not integrated classical solution in which the capacity prediction and allocation procedures are separately performed. Furthermore its performance are very near to the one of an oracle that optimizes the capacity over/under dimensioning parameter.

Application of a Long Short Term Memory neural predictor with asymmetric loss function for the resource allocation in NFV network architectures

Traffic and cloud resource prediction methodologies have been recently used in Network Function Virtualization environment for cloud and bandwidth resource allocation purposes. Both traditional and innovative prediction methodologies have been proposed for the application of allocation procedures. For instance Long Short Term Memory-based prediction techniques have been shown to be very effectiveness to allocate the resources. All of these techniques are based on the minimization of a symmetric cost function as the Root Mean Square Error that equally weights positive and negative prediction errors. However the error sign can differently impact the cost increase due to prediction errors. For instance when the Quality of Service degradation cost due to traffic loss is prevalent with respect to the cloud resource allocation cost, an algorithm is preferable that overestimates the offered traffic; conversely the traffic underestimation is preferable in the opposite case when the cloud allocation cost is higher than the QoS degradation one. For this reason we propose an Asymmetric LSTM traffic prediction procedure in which the cost function is defined so as to take into account both the QoS degradation and cloud resource allocation costs. In a typical network and traffic scenario, we show how the proposed solution allows for cost decrease by 40% with respect to classical LSTM prediction methodology based on the Root Mean Square Error.

CNSM 2019 special issue: Embracing the new wave of artificial intelligence

CNSM 2019 special issue: Embracing the new wave of artificial intelligence

A Sufficient Condition for the Existence of Restricted Fractional (g, f)-Factors in Graphs

In an NFV network, the availability of resource scheduling can be transformed to the existence of the fractional factor in the corresponding NFV network graph. Researching on the existence of special fractional factors in network structure can help to construct the NFV network with efficient application of resources. Let h: E(G) → [0, 1] be a function. We write $${d}_{G}^{h}(x)=\sum \limits_{e\ni x}h(e)$$ . We call a graph Fh with vertex set V(G) and edge set Eh a fractional (g, f)-factor of G with indicator function h if $$g(x)\le {d}_{G}^{h}(x)\le f(x)$$ holds for any x ∈ V(G), where Eh = {e : e ∈ E(G), h(e) > 0}. We say that G has property E(m, n) with respect to a fractional (g, f)-factor if for any two sets of independent edges M and N with ∣M∣ = m, ∣N∣ = n, and $$M\cap N=\varnothing $$ , G admits a fractional (g, f)-factor Fh with h(e) = 1 for any e ∈ M and h(e) = 0 for any e ∈ N. The concept of property E(m, n) with respect to a fractional (g, f)-factor corresponds to the structure of an NFV network where certain channels are occupied or damaged in some period of time. In this paper, we consider the resource scheduling problem in NFV networks using graph theory, and show a neighborhood union condition for a graph to have property E(1, n) with respect to a fractional (g, f)-factor. Furthermore, it is shown that the lower bound on the neighborhood union condition in the main result is the best possible in some sense.

Cost-Efficient Service Function Chain Orchestration for Low-Latency Applications in NFV Networks

With the popularization and development of cloud computing, network function virtualization and service function chain (SFC) provisioning have attracted increasing attention from researchers. Excellent and reliable network service is important for network development. Moreover, as the number of network users increases, network service construction costs become very high. Therefore, an efficient algorithm is necessary to provide an SFC with excellent performance and low resource costs. In this paper, we re-examine the problem of optimizing the deployment of an SFC to provide users with excellent and resource-saving network service. We propose a heuristic, closed-loop feedback (CLF) algorithm to find the shortest path to map an SFC. To solve the problem, we introduce and integrate a restricted Boltzmann machine and cross entropy to improve the performance of CLF. Simulation results demonstrate the excellent performance of CLF. The communication delay is reduced by approximately 20%, the accept ratio improves by approximately 15%, and the algorithm running time decreases by approximately 20%. In addition, the resource utilization ratio increases by approximately 15%, and the resource fragmentation ratio decreases by approximately 50%.

Adding Support for Automatic Enforcement of Security Policies in NFV Networks

Adding Support for Automatic Enforcement of Security Policies in NFV Networks

Evolution of Orchestration towards 5G

Service orchestration is an essential activity in 5G networks. It performs optimal resource allocation and provisions services in an effective sequence based on demands across a collection of physical or virtual network functions (P/VNF). This paper summarizes several orchestration environments and components along with their evolution towards 5G. A brief operational comparison of platforms such as Open Source Management and Orchestration (OSM MANO), Open Platform for NFV (OPNFV) and Open Network Automation Platform (ONAP) have been presented, along with different deployment models and architectural alternatives.

Graph-based feasibility analysis of resource dispatching in NFV networks

Graph-based feasibility analysis of resource dispatching in NFV networks

Independent set conditions for all fractional (g, f, n′, m)-critical deleted NFV networks

Independent set conditions for all fractional (g, f, n′, m)-critical deleted NFV networks