Hierarchical Control Plane Research Articles

MATE: When multi-agent Deep Reinforcement Learning meets Traffic Engineering in multi-domain networks

As the global network evolves into a large-scale interconnection system with multiple distributed domains, scalability and efficiency have become equally important for Traffic Engineering (TE). Traditional centralized TE suffers from significant computational complexity and privacy concerns, while distributed TE results in suboptimal routing decisions due to its lack of a coordination mechanism. This paper proposes MATE, an innovative, scalable, and efficient TE framework for multi-domain networks featuring a hierarchical control plane. MATE designs a three-stage workflow to optimize routing decisions for inter-domain traffic while ensuring their QoS guarantees. First, in the bottom-up topology abstraction stage, MATE aggregates network resources within each domain, creating an abstract view of the global network state while protecting each domain’s internal information. Second, in the top-down decision-making stage, MATE computes a QoS-constrained domain sequence based on the global network state and then converts it into multi-path routing in all related domains. Third, MATE conducts parallel inferences across relevant domains on traffic split ratios for multi-path routing using well-designed multi-agent reinforcement learning. We evaluate MATE on both real-world and synthetic topologies under various traffic patterns. In a large-scale topology encompassing 16 domains, MATE achieves near-optimal link utilization across 97% network scenarios, with an approximation ratio of below 1.3. The experimental results demonstrate MATE’s superiority in fulfilling QoS requirements, minimizing maximum link utilization, and maintaining robustness against traffic pattern variations and random link failures.

Multi-domain collaborative two-level DDoS detection via hybrid deep learning

In this paper, we investigate the problem of multiple network domains being threatened by Distributed Denial-of-Service (DDoS) attacks, in which a DDoS attack detection scheme is constructed based on the Software Defined Networks (SDN) hierarchical distributed control plane architecture. Specifically, we propose a two-level detection framework for collaborative DDoS attack detection in multi-domain scenarios. To detect the signs of DDoS attacks as early as possible on the attack path, a first-level coarse-grained anomaly detection method based on the Rényi entropy algorithm is proposed. The purpose is to calculate the feature entropy of normal and abnormal traffic in a simple statistical way within the local network domain, achieving rapid perception of network anomalies. Then, the root server aggregates all abnormal traffic data uploaded by each local network domain, and the DCNN-LSTM algorithm based on a hybrid deep learning model as the second-level detection method extracts the features of the suspicious traffic from both temporal and spatial dimensions to achieve fine-grained DDoS attack classification. Finally, theoretical analysis and experimental results indicate that the proposed two-level detection method in multi-domain scenarios is effective and feasible, while with high detection accuracy.

Simulation of 5G Mobile Core Network using SDN & MPLS

In particular, in order to have the best possible understanding of the effects that program-able networking has, carry out a low cost analysis of both the (software define network) system and the (multiple protocol layer switching)  system. This will give you the possible understanding of the effects that program-able networking has on the economics of a network in comparison to traditional networking, which is also known as MPLS technology. This is done so that a better understanding of the effects that program-able networking, also known as software-defined networking (SDN), technology has on the economics of a network can be gained. Conducting a quantitative investigation using an activity-based methodology to compute the (capital expenditure) and (operation expenditure) costs associated with a network was the first step in achieving this objective. This investigation was carried out.  Second, in order to compare the aforementioned architectures in terms of their financial performances, we used the USC Scalability metric in addition to the Cost-to-Service metric. Both of these metrics were used. These models include (centralized control plane), (distributed control plane), and (hierarchical control plane). This was done in order to obtain the possible understanding of the different SDN plane systems that are currently in use. The separation of control plane and data plane that is achieved through the implementation of software-defined networking (SDN) makes it possible to program networks. Because of this separation, network operators and administrators are now in a position to make better use of the network's resources and have an easier time provisioning those resources when compared to the situation before the separation took place.

MEC-Based Dynamic Controller Placement in SD-IoV: A Deep Reinforcement Learning Approach

The flow fluctuations in the highly dynamic Internet of Vehicles (IoV) make the IoV difficult to provide reliable and scalable wireless network services for the emerging applications in the 5 G and beyond era. The software-defined networks (SDN) could feasibly manage and optimize the network according to the network load. Controller placement is a critical problem in SDN to achieve its robustness and flexibility with the changes of network status. Motivated by the advantages of SDN and Mobile-edge computing (MEC), this paper aims at enhancing the performance of IoV with the assistance of these two. Specifically, we consider a three-layer hierarchical SDN control plane for the IoV, where the SDN controllers are placed at the edge of networks. Under this framework, we investigate a multi-objective optimization problem on controller placement problem including delay, load balancing, and path reliability. To efficiently solve the formulated NP-hard problems, we develop an algorithm based on multi-agent deep Q-learning networks (MADQN) because of its advantages for large-scale combinatorial optimization. At last, we use multi-process technology to accelerate the operation of the algorithm, so as to complete the algorithm iteration faster. Numerical results show that the proposed methods achieve better performances than three baselines.

Simulator for Performance Evaluation of ASON/GMPLS Network

The hierarchical control plane network architecture of Automatically Switched Optical Network with utilization of Generalized Multi-Protocol Label Switching protocols is compliant to next generation networks requirements and can supply connections with required quality of service, even with incomplete domain information. Considering connection control, connection management and network management, the controllers of this architecture could perform the same operations on the transport plane resources and could support control functions for Software Defined Networking controllers, which are considered as the future networking solutions. Therefore, it is very important to determine factors that have an influence on service control and resource management. One of the tools for achieving this goal is computer simulation. The paper proposes a simulator for the hierarchical control plane networks in an open source OMNeT++ environment, in order to evaluate network architecture performance for different network structures and traffic parameters. To demonstrate simulator capabilities, examples of typical simulations performance results for Polish and European network structures are presented and discussed. Performance metrics which have been used to evaluate the performance of this architecture are: mean values of Call Set-up Time, Connection Set-up Time, Call Release Time, Connection Release Time, loss probability for inter-domain connection requests and loss probability for low and high priority requests. General remarks concerned with presented simulator are also provided.

Quality of Service Improvement with Optimal Software-Defined Networking Controller and Control Plane Clustering

The controller is indispensable in software-defined networking (SDN). With several features, controllers monitor the network and respond promptly to dynamic changes. Their performance affects the quality-of-service (QoS) in SDN. Every controller supports a set of features. However, the support of the features may be more prominent in one controller. Moreover, a single controller leads to performance, single-point-of-failure (SPOF), and scalability problems. To overcome this, a controller with an optimum feature set must be available for SDN. Furthermore, a cluster of optimum feature set controllers will overcome an SPOF and improve the QoS in SDN. Herein, leveraging an analytical network process (ANP), we rank SDN controllers regarding their supporting features and create a hierarchical control plane based cluster (HCPC) of the highly ranked controller computed using the ANP, evaluating their performance for the OS3E topology. The results demonstrated in Mininet reveal that a HCPC environment with an optimum controller achieves an improved QoS. Moreover, the experimental results validated in Mininet show that our proposed approach surpasses the existing distributed controller clustering (DCC) schemes in terms of several performance metrics i.e., delay, jitter, throughput, load balancing, scalability and CPU (central processing unit) utilization.

An Effective Hierarchical Control Plane for Software-Defined Networks Leveraging TOPSIS for End-to-End QoS Class-Mapping

If an end-to-end (E2E) path includes multiple domains, we need inter-domain collaboration to ensure the E2E quality-of-service (QoS) for the applications. In heterogeneous networks, an E2E path may go through domains with several QoS classes in each domain. However, the prevalent legacy network architecture and the standard software-defined networking (SDN) model lack effective mechanisms for inter-domain collaboration and QoS class mapping. In this study, we propose a hierarchical SDN control plane approach to guarantee the E2E QoS among multiple domains with various QoS classes on the E2E path. We propose a controller module for selecting the most suitable QoS class for each domain in the E2E path based on multi-criteria decision-making by using the technique for order of preference by similarity to ideal solution (TOPSIS). We map the suitable service classes in the global controller (GC) for provisioning the E2E QoS according to the application service requests. First, we propose an SDN-based inter-domain communication scheme and the message processing algorithm for E2E service delivery when multiple QoS classes exist in each domain. Next, we formulate the problem of service class selection with TOPSIS, provide an E2E mapping scheme, and demonstrate it with an example. Finally, we compare the proposed approach with the existing schemes for E2E QoS class mapping in terms of E2E delay, jitter, packet loss rate (PLR), and cost (per bandwidth unit). According to our simulation results, the proposed approach ensures the E2E QoS and guarantees the E2E delay, jitter, PLR, and cost according to the application service requests.

An Analysis of Hierarchical Software-Defined Network Control Plane: A Reliability Approach

Present-day computer systems have drastically transformed from the ones in days of basic file sharing, peripheral sharing or the hosting of companywide applications on a server to much more sophisticated, small and faster systems. These systems have further expanded to include cloud-based networks, virtualized desktops, servers, etc. The capabilities of evolving heterogeneous computer systems require advanced control plane. Software-defined networking (SDN) proposes to control the network from a centralized controller instead of a distributed configuration. SDN makes it easier for network operators to evolve network capabilities. Even though SDN proposes a logically centralized system, the controllers may not represent a single, centralized device, instead the control plane may consist of logically centralized but physically distributed controllers wherein each controller manages different administrative domains of the network or different parts of the flow space. There are mainly two types of control plane architecture: flat control plane and hierarchical control plane. In this paper, we have analyzed the reliability and availability of the hierarchical SDN control plane. We take into consideration work-load capacities of the controllers, link failures, node failures and controller-end failures to determine the reliability of the system.

Performance Analysis of a Hierarchical Inter-Domain Provisioning Framework for Multi-Domain Software-Defined Optical Transport Networks

This paper analyzes the performance of a hierarchical inter-domain provisioning (HIP) framework for multi-domain software-defined optical transport networks (SD-OTNs). The challenge of this paper is to evaluate the performance and scalability of the HIP framework so that the framework can account for the complexity of the inter-domain routing and wavelength assignment (RWA) problem of the underlying OTNs. The framework is implemented in an experimental testbed to show the effectiveness of the HIP framework while ensuring seamless integration of multiple OTNs in a hierarchical control plane architecture. In addition to the experimental evaluations, the throughput, latency, and power metric of the framework based on a closed-form queueing model are evaluated analytically by considering the RWA problem of the underlying OTNs. The analytic results show that the HIP framework significantly improves the performance of a large-scale OTN with respect to the provisioning delay and throughput. Despite the marginal benefits of the blocking probability, the HIP framework permits enhanced scalability up to 24, 000 times by using a three-layer control plane architecture for multi-domain SD-OTNs.

An economic framework for analysis of network architectures: SDN and MPLS cases

As new networking architectures such as 5G start materializing in the next era of communication revolution, economic and operational facets of these future networks drive researchers to rethink about their capabilities of flexibility, agility, programmability, and cost-effectiveness. These are the main features to make networks like 5G (fifth-generation) possible and available for societies shortly because they are the fundamental catalysts to achieve economic success in a network. It is vital for network owners to analyze and understand economic aspects of a network architecture before deciding to invest in it. In this article, we introduce a framework utilizing an activity-based approach for the techno-economic analysis of network architectures. In particular, we perform an economic analysis of SDN (Software Defined Networking) technology and MPLS (Multiprotocol Label Switching) technology in order to understand how programmable networking, i.e., SDN technology, affects the network economics compared to traditional networking, i.e., MPLS technology. To this end, we firstly conduct a quantitative analysis exploiting an activity-based approach for CAPEX (Capital Expenditure) and OPEX (Operational Expenditure) calculations of a network. Secondly, we evaluate the architectures above concerning their economic performances using two metrics: Unit Service Cost Scalability metric and Cost-to-Service metric. Also, we present mathematical models to calculate certain cost parts of a network. Also, we compare different popular SDN control plane models, Centralized Control Plane (CCP), Distributed Control Plane (DCP), and Hierarchical Control Plane (HCP), to understand their economic impact with regards to the defined metrics. We use video as the service with different traffic patterns for the comparison. This work aims at being a useful primer to providing insights regarding which technology and control plane model(s) are appropriate for a specific service, i.e., video, for network owners to plan their investments.

The Fog of Things Paradigm: Road toward On-Demand Internet of Things

In this article, we introduce the concept of FoT, a paradigm for on-demand IoT. On-demand IoT is an IoT platform where heterogeneous connected things can be accessed and managed via a uniform platform based on real-time demands. Realizing such a platform faces challenges including heterogeneity, scalability, responsiveness, and robustness, due to the large-scale and complex nature of an IoT environment. The FoT paradigm features the incorporation of fog computing power, which empowers not only the IoT applications, but more importantly the scalable and efficient management of the system itself. FoT utilizes a flat-structured virtualization plane and a hierarchical control plane, both of which extend to the network edge and can be reconfigured in real time, to achieve various design goals. In addition to describing the detailed design of the FoT paradigm, we also highlight challenges and opportunities involved in the deployment, management, and operation of such an on-demand IoT platform. We hope this article can shed some light on how to build and maintain a practical and extensible control back-end to enable large-scale IoT that empowers our connected world.

Robust hierarchical control plane for Transport Software-Defined Networks

Software-Defined Networking (SDN) enables the separation of data and control planes. Today, it is in use in several practical networks. Research on SDN is yet to focus on Transport Networks in a significant way. To provide efficient services at low cost for future technologies, such as 5G and beyond, the Transport Network will need to support network functions; adapt the network to different applications' traffic; provide high bandwidth at low latencies; integrate with other networks; etc. By fulfilling these requirements, a Transport Software-Defined Network (T-SDN) can become a fundamental part of the telecommunication infrastructure. However, most current SDN solutions have been developed for Layer 3, and cannot be directly applied to a T-SDN without proper adaptations, as Transport Networks have several characteristics that are different from those of Layer 3 networks. In particular, the design of a T-SDN control plane must address heterogeneity in terms of protocols and administrative network areas; as well as high reliability. A hierarchical control plane is suitable to support these characteristics. Accordingly, in this study, we analyze how to design a robust hierarchical control plane for T-SDNs. We discuss how resiliency against random failures can be provided through redundancy; and how survivability against correlated failures (such as disasters) can be achieved by effectively choosing network nodes where to place control-plane elements, and deciding how to route control-plane traffic. We formulate an Integer Linear Program to design a hierarchical, failure- and disaster-resilient T-SDN control plane. We also propose a heuristic for post-failure switch-controller reassignment. We compare our model with a disaster-unaware control-plane design whose objective is to reduce network-resource utilization. Our results show that we can achieve much higher disaster and failure resiliency, at the cost of slightly larger network-resource utilization.

Topology-Preserving Traffic Engineering for Hierarchical Multi-Domain SDN

Decoupling of control and data plane in Software Define Networks (SDN) creates significant flexibility in network management. As networks are evolving into a complex multi-domain multi-layer architecture, traffic engineering across multiple domains and layers entails significant challenges for the control plane, especially when each separate administrative domain is not willing to disclose its network topology and resource information. In this paper, we present the first traffic engineering scheme for a multi-domain SDN by using a hierarchical control plane, in which each domain is managed by a local controller, and they are further controlled by a centralized root controller. To preserve the local topology of each domain, we extend the Network Information Base (NIB) into a hierarchical architecture as well, and the root controller maintains only an abstracted view of the lower-layer networks in its NIB. Then, we devise a communication protocol, which enables controllers at different layers and domains to work collaboratively on bandwidth allocation by reading to the hierarchical NIB. We also present an improved traffic engineering algorithm by considering bandwidth and delay simultaneously, to maximize the network utilization while respecting max-min fairness. Experiments on a 719-switches 5-domain network topology demonstrate that our proposal could drive the link utilization ratio to more than 85%.

Economic Viability of Software Defined Networking (SDN)

Economical and operational facets of networks drive the necessity for significant changes towards fundamentals of networking architectures. Recently, the momentum of programmable networking attempts illustrates the significance of economic aspects of network technologies. Software Defined Networking (SDN) has got the attention of researchers from both academia and industry as a means to decrease network costs and generate revenue for service providers due to features it promises in networking. In this article, we investigate how programmable network architectures, i.e. SDN technology, affect the network economics compared to traditional network architectures, i.e. MPLS technology. We define two metrics, Unit Service Cost Scalability and Cost-to-Service, to evaluate how SDN architecture performs compared to MPLS architecture. Also, we present mathematical models to calculate certain cost parts of a network. In addition, we compare different popular SDN control plane models, Centralized Control Plane (CCP), Distributed Control Plane (DCP), and Hierarchical Control Plane (HCP), to understand the economic impact of them with regards to the defined metrics. We use video traffic with different patterns for the comparison. This work aims at being a useful primer to providing insights regarding which technology and control plane model are appropriate for a specific service, i.e. video, for network owners to plan their investments.

Orchestration of RAN and Transport Networks for 5G: An SDN Approach

The fifth generation of mobile networks is planned to be commercially available in a few years. The scope of 5G goes beyond introducing new radio interfaces, and will include new services like low-latency industrial applications, as well as new deployment models such as cooperative cells and densification through small cells. An efficient realization of these new features greatly benefit from tight coordination among radio and transport network resources, something that is missing in current networks. In this article, we first present an overview of the benefits and technical requirements of resource coordination across radio and transport networks in the context of 5G. Then, we discuss how SDN principles can bring programmability to both the transport and radio domains, which in turn enables the design of a hierarchical, modular, and programmable control and orchestration plane across the domains. Finally, we introduce two use cases of SDN-based transport and RAN orchestration, and present an experimental implementation of them in a testbed in our lab, which confirms the feasibility and benefits of the proposed orchestration.

Load-Balancing Multiple Controllers Mechanism for Software-Defined Networking

Software-Defined Networking (SDN) is a new form for networking that has the potential to have a major impact on Internet technology. Key aspects of SDN include the separation of data plane and the control plane. SDN architecture facilitates the management of complex networks and represents a new solution for future network applications Internet technology. As the size of networks increases, the scalability of the centralized controller becomes increasingly important. This work proposed a load-balancing mechanism for use in a multiple-controller SDN environment by implement a hierarchical control plane with both a meta-control plane and a local control plane. The meta-control plane analyzes the resources and utilization of the local control plane to optimize processing performance. This mechanism supports the load balancing of the local control plane to optimize data plane performance and eliminate the bottleneck of the centralized control in a network. This work analyzes the proposed load-balancing mechanism in a multiple-controller SDN environment. The results indicate that the meta controller based manager mechanism can monitors and effectively deal with the loading of the overloaded local controller. Based on the results thus obtained, the proposed local control plane scheduling balances the loadings of the multiple controllers and increases the network throughput by 12.7 and 9.2 % over those achieved using a centralized controller and multiple controllers without a scheduling mechanism.

Demonstration of Dynamic Resource Sharing Benefits in an Optical C-RAN

The next generation of mobile communication (i.e., 5G) will bring new challenges for the transport infrastructure, e.g., in terms of flexibility and capacity. The joint orchestration of radio and transport resources can help to address some of these challenges. One example is the possibility of reconfiguring the use of the transport network resources according to the spatial and temporal variations of the wireless traffic patterns. Using the concept of dynamic resource sharing, a limited pool of transport resources can be shared among a large number of radio base stations (RBSs), thus reducing considerably the overall deployment cost of the transport infrastructure. This paper proposes a provisioning strategy for a centralized radio access network with an optical transport whose wavelength resources can be dynamically shared among multiple RBSs. The proposed strategy utilizes a hierarchical software-defined networking control plane where a global orchestrator optimizes the usage of radio and transport resources. The benefits of the proposed strategy are assessed both by simulation and by experiment via an optical data plane emulator developed for this purpose. It is shown that the dynamic resource sharing can save up to 31.4% of transport resources compared to a conventional dimensioning approach, i.e., based on the overprovi-sioning of wavelength resources.

Survivable Virtual Topology Design in Multi-domain Optical Networks

We consider the problem of survivable virtual network mapping in a multi-domain optical network with the objective of minimizing total network link cost for a given virtual traffic demand. The survivability constraint guarantees the connectivity of virtual nodes after any single optical link failure. We propose a hierarchical software-defined networking-based control plane to exchange information between domains, and we propose heuristic approaches for mapping virtual links onto multi-domain optical links using partition and contraction mechanisms on the virtual topology. We provide an integer linear programming formulation to compare with our heuristic approaches. Numerical results show that our heuristic approach is effective in reducing total network cost and increasing the successful mapping rate.

An intelligent inter-domain routing scheme under the consideration of diffserv QoS and energy saving in multi-domain software-defined flexible optical networks

Large scale multi-domain software-defined optical networks (SDON) provisioning has become a key area with increased scalable bandwidth services across wider network regions. Although distributed schemes could achieve lightpath routing by the ergodic process of domain boundary nodes, it increases the complexity of the signaling procedure and deployment of the interface. Moreover, the physical impairments are always the main factor of the infrastructure layer in SDON, which affect the transmission quality of the signal. Meanwhile, with increasing energy consumption of the Internet, it is imperative to design energy-efficient networks. To address the above issues, in this paper, an intelligent inter-domain routing scheme, which is supported by hierarchical control plane architecture, is presented based on sub-topology graph under the consideration of diffserv quality-of-service (QoS) and energy saving. The proposed scheme could achieve multi-domain quality of transmission (QoT), energy aware routing and spectrum assignment (RSA). We explore the scenarios where the multi-domain SDON achieve energy efficiency on the basis of meeting the QoT requirement. The blocking, energy consumption and average set up delay performances of the proposed schemes are estimated. The results indicate that the introduction of sub-topology in multi-domain RSA scheme has the better performance comparing with the distributed scheme.

A Hybrid Hierarchical Control Plane for Flow-Based Large-Scale Software-Defined Networks

The decoupled architecture and the fine-grained flow-control feature limit the scalability of a flow-based software-defined network (SDN). In order to address this problem, some studies construct a flat control plane architecture; others build a hierarchical control plane architecture to improve the scalability of an SDN. However, the two kinds of structure still have unresolved issues: A flat control plane structure cannot solve the superlinear computational complexity growth of the control plane when the SDN scales to a large size, and the centralized abstracted hierarchical control plane structure brings a path stretch problem. To address these two issues, we propose Orion, a hybrid hierarchical control plane for large-scale networks. Orion can effectively reduce the computational complexity of an SDN control plane by several orders of magnitude. We also design an abstracted hierarchical routing method to solve the path stretch problem. Furthermore, we propose a hierarchical fast reroute method to illustrate how to achieve fast rerouting in the proposed hybrid hierarchical control plane. Orion is implemented to verify the feasibility of the hybrid hierarchical approach. Finally, we verify the effectiveness of Orion from both the theoretical and experimental aspects.