Octans: Optimal Placement of Service Function Chains in Many-Core Systems

Network function virtualization (NFV) brings the potential to provide the flexible implementation of network functions and reduce overall hardware cost by running service function chains (SFCs) on commercial off-the-shelf servers with many-core processors. Towards this direction, both academia and industry have spent vast amounts of effort to address the optimal placement challenges of NFV middleboxes. Most of the servers usually are equipped with Intel X86 processors, which adopt Non-Uniform Memory Access (NUMA) architecture. However, existing solutions for placing SFCs in one server either ignore the impact of hardware architecture or overlook the dependency between middleboxes. Our empirical analysis shows that the placement of virtual network functions (VNFs) with interdependency in a server needs more particular consideration. In this paper, we first manage the optimal placement of VNFs by jointly considering the discrepancy of cores in different NUMA nodes and interdependency between network functions (NFs), and formulate the optimization problem as a Non-Linear Integer Programming (NLIP) model. Then we find a reasonable metric to describe the dependency relation formally. Finally, we propose a heuristic-based backtracking placement algorithm (BPA) to find the near-optimal placement solution. The evaluation shows that, compared with two state-of-art placement strategies, our algorithm can improve the aggregate performance by an average of 20% or 45% within an acceptable time range.

Efficient service reconfiguration with partial virtual network function migration

Service Function Chaining (SFC) serves the traffic of a specific service along an ordered set of Service Functions (SFs). SFC uses Software Defined Networking (SDN) and Network Function Virtualization (NFV) technologies to reach the deployment and removal of SFC in an appropriate time with minimal costs. However, during the life time of deployed SFCs, the SFs are exposed to the risk of overloading, which results in an end-to-end packets drop and high delay at the application level. This paper presents the concept of a Scalable SFC Orchestrator capable of deploying SF Chains following the ETSI NFV architectural model, as well as orchestrating the runtime phase by rerouting the traffic to a different path in case of overload of certain SF instances. Moreover, a selection algorithm is presented to build the updated paths after scaling. The Scalable Orchestration feature is implemented and added as a main feature to an SFC Orchestrator. Furthermore, it is integrated into a current NFV architecture and evaluated in an NFV environment making use of the Fraunhofer FOKUS Open Baton toolkit in an OpenStack and Open Daylight based environment.

The increasing demand in Mobile Edge Computing (MEC) networks has led to the emergence of Network Function Virtualization (NFV) technology. Through the virtualization technique, NFV can decouple each Virtual Network Function (VNF) from the hardware and deploy it flexibly in the MEC. Each service request in NFV assembles a Service Function Chain (SFC) by tying different VNFs and processes the traffic in the chained order. How to allocate resources and deploy SFC in MEC to meet Quality of Service (QoS) requirements is an important challenge for NFV. An increase in the number of VNFs with traffic passing one after the other causes high delay in SFC assembly. Also, the demanded resources may change during SFC assembly, which is clearly neglected in existing works. Therefore, high delay and resource demand uncertainty during SFC assembly affect QoS. With this motivation, we solve the SFC deployment problem by Parallelizing VNFs under Resource Demand Uncertainty (PVRDU) in MEC. We use the dependency between VNFs to transform the original SFC into a parallel SFC and then assemble several sub-SFCs based on a Deep Reinforcement Learning (DRL) approach. Meanwhile, PVRDU uses a Markov-based approximation algorithm to handle resource demand uncertainty. We perform extensive trace-driven simulations to verify the effectiveness of the proposed algorithm. The evaluation results of PVRDU are promising in different aspects compared to state-of-the-art methods. Specifically, for the considered scenario, DPC reduced SFC delay by as much as 8.7%.

Network function virtualization (NFV) is a key networking concept whose benefits include scalability, flexibility, and cost-effective service provisioning. In NFV, service function chains (SFCs) adaptable to customers' needs are created by chaining virtualized network functions (VNFs). VNFs and SFCs are sensitive elements that, if compromised, would affect network security. The detection of compromised VNFs and SFCs is imperative, and although anomaly detection can be used in such a context, there is a lack of research work on the use of anomaly detection in NFV. In this article, we exploit the use of anomaly detection mechanisms to identify suspicious VNFs and SFCs. We introduce, into a widely accepted NFV architecture, an NFV security module (NSM) that, by analyzing VNFs and SFCs' operations, detects anomalies possibly resulting from security attacks. To prove the concept, three mechanisms have been implemented and deployed in NSM to observe how anomaly detection performs, given quantitative and qualitative information. We found out that anomaly detection is effective for VNF and SFC security and, in the case of using entropy as anomaly detection technique, it presents accuracy of up to 98 percent without harming NFV environment operations.

Network Slicing (NS) technique is comprehensively reshaping the next-generation communication networks (e.g. 5G, 6G). Software-Defined Networking (SDN) and Network Functions Virtualization (NFV) predominantly control the flow of service functions on NS to incorporate versatile applications as per user demands. In the virtualized-Software Defined Networking vSDN environment, a chain of well-defined virtual network functions (VNFs) are installed on Service Function Chains (SFCs) by multiple Internet Service Providers (ISPs) concurrently. Generation, allocation, re-allocation, release and destroying associative VNFs on SFC is an extremely difficult task while keeping high selection accuracy. Towards solving this fundamental issue, in this work, we have proposed a multi-layered SFC formation for adaptive VNF allocation on dynamic slices. We have formulated an ILP to address the VNF-EAP (VNF-Embedding and Allocation Problem) over real network topology (AT&T Topology). Leveraging machine learning techniques we have shown an intelligent VNF selection mechanism to optimize resource utilization. The performance evaluation shows remarkable efficiency on ML-driven dynamic VNF selections over static allocations on SFCs by halving resource usage. Further, we have also studied a VNF typecasting technique for service backup on outage slices in the field of disaster management activities.

In this paper, we propose the use of reinforcement learning to deploy a service function chain (SFC) of cellular network service and manage the VNFs operation. We consider that the SFC is deployed by the reinforcement learning agent considering a scenario with distributed data centers, where the virtual network functions (VNFs) are deployed in virtual machines in commodity servers. The VNF management is related to create, delete, and restart the VNFs. The main purpose is to reduce the number of lost packets taking into account the energy consumption of the servers. We use the Proximal Policy Optimization (PPO2) algorithm to implement the agent and preliminary results show that the agent is able to allocate the SFC and manage the VNFs, reducing the number of lost packets.

The modern network is equipped with many service functions to acquire high-quality service. The emergence of network function virtualization (NFV) provides a convenient way to accomplish the network services in the form of virtual network function (VNF) and also makes the scheduling solution of VNFs flexible. The VNFs can be deployed on commodity servers as software processes. Besides, multiple VNFs are chained in a specified order as a service function chain (SFC) to serve a given flow, increasing the scheduling difficulty to minimize the average flow completion time. In this paper, we study the problem of scheduling multiple SFCs with the constraint of different resource limitations in various commodity servers. This problem is typically formulated as an Integer Linear Programming (ILP) problem, which is NP-hard. To well tackle this problem, we propose a deep reinforcement learning (DRL) approach. It involves multi-step decision making, which can be naturally transformed into a DRL problem. We design specific reward and state representations for such a multi-resource scheduling problem. We also consider how to use DRL to handle online requests of SFCs. The experiment results demonstrate that the DRL approach can significantly reduce the average completion time of a set of SFC and achieves a cost saving of 39.94% against the benchmark method.

Network Function Virtualization (NFV) technology facilitates flexible and fast service provisioning by deploying network functions in wished positions and chaining order as Service Function Chain (SFC). Empirically, if the processing inside some Virtual Network Functions (VNFs) does not relate to each other, parallel chaining them can further improve the performance, such as reducing the end-to-end hops of SFCs. However, it is often ignored that parallel deploy VNFs will also introduce additional VNFs to replicate, distribute, and merge data. This will add hops and increase resource consumption. Will parallel deployment of VNFs in SFCs necessarily bring benefits? Regarding multiple parallel forms of an SFC, which one is the best? Further, how does the service process of the NFV network affect the parallel decisions of VNFs? In this paper, we design an NFV orchestrator, DP-NFVO, to maximize the revenue of NFV Service Providers and provide quantitative insights into the impact of parallel VNFs deploy-ment. Precisely, we first refine existing VNF parallelism rules and discuss the impact of VNF parallelism on latency and resource consumption, and analyze the optimal form of parallelism for SFCs. Second, we decompose the service request process in NFV networks into multiple stages and provide a joint model of VNF parallelism and deployment. Further, we design a VNF online deployment and parallelism decision policy for DP-NFVO based on Proximal Policy Optimization (PPO) to maximize revenue. Finally, we demonstrate the effectiveness and adaptability of DP-NFVO on parallel processing and the requests service process with extensive experiments.

A comprehensive survey of service function chain provisioning approaches in SDN and NFV architecture

Service Function Chain (SFC) integrates Network Function Virtualization (NFV) for flexible deployment of Virtualized Network Function (VNF) and utilizes Software-Defined Networking (SDN) for establishing dynamic network connections, enabling customized service chains for specific applications. Security risks in NFV/SDN networks directly impact SFC, making them vulnerable to attacks. Traditional security solutions rely on passive defense mechanisms, which are insufficient against the uncertainties brought by vulnerabilities and backdoors in open networks. As an emerging network security technology, Moving Target Defense (MTD) is expected to break through traditional security protection mechanisms such as “enhanced”, “plug-in” and “passive” defense to achieve active defense against both known and unknown threats. Features of NFV/SDN networks, such as dynamic resource scheduling, microservices, and centralized control, also offer critical guidance for implementing MTD strategy. Therefore, in this paper, we propose a novel security enhancement mechanism by equipping SFC and VNFs with MTD to inherently protect SFC and VNFs during the deployment process, using a sub-pool partitioning algorithm to enhance heterogeneity across sub-pools. Additionally, to address Quality of Service (QoS) requirements, we formulate an optimization problem with three criteria: latency, cost, and defense success rate. Following that, we propose a dynamic deployment algorithm based on Deep Reinforcement Learning (DRL). Finally, to validate the effectiveness of the proposed algorithm, we conduct extensive experiment which have demonstrated that our proposed algorithm is good in increasing the security gains of both SFC and VNFs, provided that this improvement comes at the expense of modest cost and latency.

Service function chaining (SFC), based on the network function virtualization (NFV) technology, is a promising candidate for realizing agile and elastic network service provisioning. This paper tackles a SFC deployment across multiple domains where only abstracted intra-domain resource information is available. For the multi-domain SFC deployment, a privacy preserving deployment mechanism called PPDM has been proposed. To conceal privacy of domains, PPDM abstracts intradomain resource information to a binary matrix, and adopts deep reinforcement learning (DRL) for flexible SFC deployment. However, because PPDM deploys all Virtual Network Functions (VNFs) included in a single SFC in a batch manner, substrate nodes need to redundantly reserve the resources required by the deployable VNFs until the construction of the SFC is completed. As a result, the deployment of the SFC may fail because of the lack of the resources on substrate nodes. In this paper, to improve acceptance ratio of SFC requests, we propose a privacy-preserving sequential deployment mechanism (PPSDM) for SFC deployment across multiple domains. The key idea behind PPSDM is to deploy VNFs in a single SFC one by one in a sequential manner to avoid redundant reservation of resources. Comprehensive simulations demonstrate that PPSDM achieves up to 22% higher acceptance ratio of SFC requests compared to the baseline approaches such as PPDM.

Network Functions Virtualization (NFV) is an emerging initiative where virtualization is used to consolidate Network Functions (NFs) onto high volume servers (HVS), switches, and storage. In addition, NFV provides flexibility as Virtual Network Functions (VNFs) can be moved to different locations in the network. One of the major challenges of NFV is the allocation of demanded network services in the network infrastructures, commonly referred to as the Network Functions Virtualization - Resource Allocation (NFV-RA) problem. NFV-RA is divided into three stages: (i) Service Function Chain (SFC) composition, (ii) SFC embedding and (iii) SFC scheduling. Up to now, existing NFV-RA approaches have mostly tackled the SFC embedding stage taking the SFC composition as an assumption. Few approaches have faced the composition of the SFCs using heuristic approaches that do not guarantee optimal solutions. In this paper, we solve the first stage of the problem by characterizing the service requests in terms of NFs and optimally building the SFC using an Integer Linear Programming (ILP) approach.

With the emergence of Network Function Virtualization (NFV) and Software-Defined Networks (SDN), Service Function Chaining (SFC) has evolved into a popular paradigm for carrying and fulfilling network services. However, the implementation of Mobile Edge Computing (MEC) in sixth-generation (6G) mobile networks requires efficient resource allocation mechanisms to migrate virtual network functions (VNFs). Deep learning is a promising approach to address this problem. Currently, research on VNF migration mainly focuses on how to migrate a single VNF while ignoring the VNF sharing and concurrent migration. Moreover, most existing VNF migration algorithms are complex, unscalable, and time-inefficient. This paper assumes that each placed VNF can serve multiple SFCs. We focus on selecting the best migration location for concurrently migrating VNF instances based on actual network conditions. First, we formulate the VNF migration problem as an optimization model whose goal is to minimize the end-to-end delay of all influenced SFCs while guaranteeing network load balance after migration. Next, we design a Deep Learning-based Two-Stage Algorithm (DLTSA) to solve the VNF migration problem. Finally, we combine previous experimental data to generate realistic VNF traffic patterns and evaluate the algorithm. Simulation results show that the SFC delay after migration calculated by DLTSA is close to the optimal results and much lower than the benchmarks. In addition, it effectively guarantees the load balancing of the network after migration.

To reduce the processing delay from the sequentially running virtual network functions (VNFs) in a service function chain (SFC), network function parallelism (NFP) is introduced that allows VNFs of the SFC to run in parallel. Existing NFP solutions only focused on improving parallelism benefits without paying much attention to resource utilization while deploying VNFs of SFCs. We take advantage of resource-delay dependency to propose a flexible and efficient parallelized SFC placement mechanism called FlexSFC which determines the optimal SFC placement while reducing resource usage and meeting end-to-end delay guarantees of the SFCs deployed. Initial results show that FlexSFC guarantees the end-to-end delay requirement with better resource utilization and SFC acceptance rate than the state-of-the-art approaches.