Network Virtualization Techniques Research Articles

Distributed Orchestration of Service Function Chains for Edge Intelligence in the Industrial Internet of Things

Network virtualization techniques are promising to overcome the obstacle of applying and expanding costly traditional networks in the industrial Internet of things (IIoT). Artificial intelligence (AI)-enhanced distributed resource management in edge networks has aroused researchers’ widespread attention. However, dynamically arrived service requests and limited edge resources complicate the service scheduling issue. In this article, we establish a dynamic network virtualization technique enabled service function chain (SFC) orchestration framework in IIoT, formulate the joint optimization problem to maximize total utility and decompose it into two subproblems, i.e., SFC selection and dynamic SFC orchestration. A dynamic orchestration of SFC (DOS) scheme, consisting of resource-aware matching algorithm and averaged multistep double deep q-network algorithm, is designed to embed SFC requests distributedly on the optimal virtualized network function chains. At last, we validate the superiority of our proposed DOS scheme by experimental results.

A review of edge computing: Features and resource virtualization

With the advent of Internet of Things (IoT) connecting billions of mobile and stationary devices to serve real-time applications, cloud computing paradigms face some significant challenges such as high latency and jitter, non-supportive location-awareness and mobility, and non-adaptive communication types. To address these challenges, edge computing paradigms, namely Fog Computing (FC), Mobile Edge Computing (MEC) and Cloudlet, have emerged to shift the digital services from centralized cloud computing to computing at edges. In this article, we analyze cloud and edge computing paradigms from features and pillars perspectives to identify the key motivators of the transitions from one type of virtualized computing paradigm to another one. We then focus on computing and network virtualization techniques as the essence of all these paradigms, and delineate why virtualization features, resource richness and application requirements are the primary factors for the selection of virtualization types in IoT frameworks. Based on these features, we compare the state-of-the-art research studies in the IoT domain. We finally investigate the deployment of virtualized computing and networking resources from performance perspective in an edge-cloud environment, followed by mapping of the existing work to the provided taxonomy for this research domain. The lessons from the reviewed are that the selection of virtualization technique, placement and migration of virtualized resources rely on the requirements of IoT services (i.e., latency, scalability, mobility, multi-tenancy, privacy, and security). As a result, there is a need for prioritizing the requirements, integrating different virtualization techniques, and exploiting a hierarchical edge-cloud architecture.

A Continuous-Decision Virtual Network Embedding Scheme Relying on Reinforcement Learning

Network Virtualization (NV) techniques allow multiple virtual network requests to beneficially share resources on the same substrate network, such as node computational resources and link bandwidth. As the most famous family member of NV techniques, virtual network embedding is capable of efficiently allocating the limited network resources to the users on the same substrate network. However, traditional heuristic virtual network embedding algorithms generally follow a static operating mechanism, which cannot adapt well to the dynamic network structures and environments, resulting in inferior nodes ranking and embedding strategies. Some reinforcement learning aided embedding algorithms have been conceived to dynamically update the decision-making strategies, while the node embedding of the same request is discretized and its continuity is ignored. To address this problem, a Continuous-Decision virtual network embedding scheme relying on Reinforcement Learning (CDRL) is proposed in our paper, which regards the node embedding of the same request as a time-series problem formulated by the classic seq2seq model. Moreover, two traditional heuristic embedding algorithms as well as the classic reinforcement learning aided embedding algorithm are used for benchmarking our prpposed CDRL algorithm. Finally, simulation results show that our proposed algorithm is superior to the other three algorithms in terms of long-term average revenue, revenue to cost and acceptance ratio.

A comprehensive and systematic review of the network virtualization techniques in the IoT

SummaryInternet of Things (IoT) as one of the last technology in the network domain is an environment of associated physical objects that are reachable through the Internet. Virtualization as the primary technology of IoT has a significant role in its popularization. The visualization strategies have a great impact on the IoT, but, as far as we know, there is not a comprehensive and systematic study in this field. Also, there is no systematic and complete way to discuss and analyze the related strategies. Henceforth, a systematic survey of virtualization techniques is presented in this paper. Different parameters based on the examination of the prevailing methods are considered in three main categorizations. Furthermore, the benefits and weaknesses connected with selected techniques have been discussed, and the significant issues of these techniques are addressed to improve the more efficient virtualization technique for the future.

Satellite Mobile Edge Computing: Improving QoS of High-Speed Satellite-Terrestrial Networks Using Edge Computing Techniques

The high-speed satellite-terrestrial network (STN) is an indispensable alternative in future mobile communication systems. In this article, we first introduce the architecture and application scenarios of STNs, and then investigate possible ways to implement mobile edge computing (MEC) technique for QoS improvement in STNs. We propose satellite MEC (SMEC), in which a user equipment without a proximal MEC server can also enjoy MEC services via satellite links. We propose a dynamic network virtualization technique to integrate the network resources, and furtherly design a cooperative computation offloading (CCO) model to achieve parallel computation in STNs. Task scheduling models in SMEC are discussed in detail, and an elemental simulation is conducted to evaluate the performance of the proposed CCO model in SMEC.

A Multi-Objective Virtual Network Embedding Algorithm in Cloud Computing

Service-oriented network virtualization enables a Network-as-a-Service paradigm to be available in cloud computing, thereby allowing network infrastructures to be exposed and utilized on demand. Virtual network embedding is a core network virtualization technique; however, a number of conflicting embedding objectives must be accommodated when implementing virtual network embedding. In this paper, we propose a multi-objective virtual network embedding framework to balance various mapping (or embedding) objectives. In addition, we apply an improved multi-objective group search optimizer to solve the virtual network embedding problem and generate trade-off solutions. Experimental results suggest that multi-objective virtual network embedding can balance various mapping objectives while achieving excellent results for acceptance ratios and revenue.

Study of Reconfiguration Cost and Energy Aware VNE Policies in Cycle-Stationary Traffic Scenarios

Network virtualization techniques allow for the coexistence of many virtual networks hosted in the same substrate network. Virtual router migration allows for resource consolidation with the consequence to reduce the power consumption in low traffic periods. Unfortunately, virtual router migration has the effect of degrading the quality of service during the downtime in which the router is not able to carry on its forwarding function. For this reason, despite the advantages in power consumption saving, the migration technique should be applied with parsimony to avoid an excessive QoS degradation. The objective of this paper is to study virtual network embedding (VNE) problems aware of both operation and reconfiguration costs that are characterized by the energy consumption and the revenue loss due to QoS degradation. Both embedding optimization problem formulation and feasible solutions to the problem in cycle-stationary traffic scenario and when the possible embeddings are determined a priori will be given. Two migration policies will be introduced and compared. The first one, referred to as “global,” is based on the knowledge of the entire daily traffic profile and it is determined by solving a Markov decision process. The second one, referred to as “local,” is only based on the knowledge of the current traffic. The results achieved show how the application of the global policy allows better performance than the local one and its application can lead to a cost reduction in the order of 35% with respect to traditional migration policies.

A technique for full flow virtualization of multi-tenant OpenFlow networks

OpenFlow provides new capabilities for wide-area network applications such as traffic engineering and security applications with many implementations having been proposed. OpenFlow allows flows to be described using arbitrary values or wildcards in the header fields. This functionality enables a controller to manage flow entries using only a limited number of the entries. In particular, in the case of wide-area networks, the sharing of the infrastructure among tenants (e.g., application service providers and their customers) is necessary, owing to the high capital outlay and operational costs. A technique for multi-tenancy creates logically isolated virtual OpenFlow networks so that multiple tenants can handle flows independently on a single physical OpenFlow network. Of the proposed techniques, OpenVirteX virtualizes the header address space (called the flow space) for tenants. This encourages tenants to participate because, differing from some other implementations, negotiations by tenants to divide the flow space are not required. However, OpenVirteX forces tenant controllers to set either exact Media Access Control (MAC) or Internet Protocol (IP) address in the matching fields because matching fields in the physical OpenFlow switches must have either exact MAC or IP addresses to isolate flows of the different tenants. Although prefix or wildcard matching is widely used to handle a large number of flows with a small number of flow setting by an OpenFlow controller, there is no available virtualization technique that enables prefix or wildcard matching with flow space virtualization. In this paper, we propose an OpenFlow network virtualization technique that allows prefix or wildcard matching with flow space virtualization. The proposed technique translates flow entries set by tenant controllers using prefix or wildcard matching to exact-match flow entries through lazy evaluation of the actual data packets. Evaluation results obtained using a prototype of the proposed technique show that the overhead of virtualization has an acceptable impact.

Virtualization-based Cognitive Radio Networks

The emerging network virtualization technique is considered as a promising technology that enables the deployment of multiple virtual networks over a single physical network. These virtual networks are allowed to share the set of available resources in order to provide different services to their intended users. While several previous studies have focused on wired network virtualization, the field of wireless network virtualization is not well investigated. One of the promising wireless technologies is the Cognitive Radio (CR) technology that aims to handle the spectrum scarcity problem through efficient Dynamic Spectrum Access (DSA). In this paper, we propose to incorporate virtualization concepts into CR Networks (CRNs) to improve their performance. We start by explaining how the concept of multilayer hypervisors can be used within a CRN cell to manage its resources more efficiently by allowing the CR Base Station (BS) to delegate some of its management responsibilities to the CR users. By reducing the CRN users’ reliance on the CRN BS, the amount of control messages can be decreased leading to reduced delay and improved throughput. Moreover, the proposed framework allows CRNs to better utilize its resources and support higher traffic loads which is in accordance with the recent technological advances that enable the Customer-Premises Equipments (CPEs) of potential CR users (such as smart phone users) to concurrently run multiple applications each generating its own traffic. We then show how our framework can be extended to handle multi-cell CRNs. Such an extension requires addressing the self-coexistence problem. To this end, we use a traffic load aware channel distribution algorithm. Through simulations, we show that our proposed framework can significantly enhance the CRN performance in terms of blocking probability and network throughput with different primary user level of activities.

WiFi Network Virtualization to Control the Connectivity of a Target Service

This paper proposes a WiFi network virtualization technique to control the connectivity of a target service. The packet-level delay violation ratio can be reduced even in a congested situation by provisioning dedicated base station (BS) resources (a set of dedicated BSs) to the target service and allowing only the corresponding terminals to associate with the BSs. The proposed technique is novel in that BSs are specially configured to use the same MAC address, and thus all the decisions on BS selection and handover are separated from those BSs and terminals and are put together into a centralized controller, while consistent layer-2 data paths in a backhaul network are also cooperatively configured. Simulation results show that the proposed technique can control the delay violation ratio of a target VoIP service and keep the ratio extremely low and comparable to that under IEEE 802.11e. A proof-of-concept prototype including two multi-channel virtualization-capable WiFi BSs and a BS switch is developed using off-the-shelf WiFi modules and a commercial OpenFlow switch. Experimental results show that the terminals can make a handover to a dedicated BS in less than 65 ms without any packet drop and association break, and confirm that the effect of the managed handover is limited even in a VoIP application.

Network Virtualization for QoS-Aware Resource Management in Cloud Data Centers: A Survey

The increasing popularity of Cloud Computing is leading to the emergence of large virtualized data centers hosting increasingly complex and dynamic IT systems and services. Over the past decade, the efficient sharing of computational resources through virtualization has been subject to intensive research, while network management in cloud data centers has received less attention. A variety of network-intensive applications require QoS (Quality-of-Service) provisioning, performance isolation and support for flexible and efficient migration of virtual machines. In this paper, we survey existing network virtualization approaches and evaluate the extent to which they can be used as a basis for realizing the mentioned requirements in a cloud data center. More specifically, we identify generic network virtualization techniques, characterize them according to their features related to QoS management and performance isolation, and show how they can be composed together and used as building blocks for complex network virtualization solutions. We then present an overview of selected representative cloud platforms and show how they leverage the generic techniques as a basis for network resource management. Finally, we outline open issues and research challenges in the area of performance modeling and proactive resource management of virtualized data center infrastructures.

Multicast virtual network mapping for supporting multiple description coding-based video applications

Video applications have gradually become one of the mainstreams of Internet traffic along with the technological advances in recent years. In parallel, network virtualization technique allows Internet Service Providers (ISPs) to be decoupled from the physical network infrastructure operators to provide various video services to customers through tailored Virtual Networks (VNs). In a video multicast VN, the ISP could appropriately determine and configure the virtual nodes and links to deliver customized video service to the affiliated end users. However, available multicast tree mapping algorithms are based on the strong assumption that the locations of source and a collection of destination nodes are known a priori. Also, they have not taken any intermediate nodes between source and destinations into the VN mapping request. These assumptions greatly limit the adoption of advanced video coding techniques. In this paper, the multicast networks mapping for enabling Multiple Description Coding (MDC) based video applications is explored. A novel multicast mapping algorithm, MMPC, is proposed based on path convergence to flexibly identify and configure nodes (with automatic intermediate nodes identification) in multicast trees to meet certain criteria. The suggested approach is described in details and its key characteristics are analyzed through a set of analytical proves. Extensive simulation experiments are carried out to assess its performance (e.g. acceptance ratio, mapping cost and mapping time) against a range of multicast VN request scenarios. The simulation results demonstrate the superiority of MMPC algorithm in terms of multicast mapping efficiency and cost by using the conventional two-phase VN mapping algorithm as a comparison benchmark.

A two-dimensional architecture for end-to-end resource management in virtual network environments

In recent years, various network virtualization techniques have been proposed for flexibly supporting heterogeneous services over virtual network platforms. However, systematic views on how virtual network resources (VNRs) can be practically managed in such open environments has been missing till now. To fill the gap, we present in this article a two-dimensional architecture for end-to-end VNR management from distinct viewpoints of service providers and network providers. The horizontal dimension of VNR management allows SPs to bind VNRs rented from heterogeneous NPs to form unified end-to-end service delivery platforms. The vertical dimension of VNR management enables NPs to perform cost-efficient allocation of VNRs to requesting SPs, but without necessarily forcing themselves to collaborate with each other. Such a VNR management architecture will complement existing network virtualization platforms in accelerating the realization of virtual resource sharing in the future Internet business marketplaces.

Analytical modeling of context-based multi-virtual wireless mesh networks

The flexibility and self-* properties of wireless mesh networks (WMNs) and the programmable management of network resources brought by the innovative network virtualization techniques, are a twofold force to enable personalized access over wireless environments. Therefore, we design a context-aware multi-virtual architecture for WMNs to deal with the requirements of mesh clients and their applications (context can be defined as a set of requirements such as cost, security, mobility, applications’ Quality-of-Service - QoS). In this approach, a WMN is split into several adaptable Virtual Networks (VNs), each one appropriate to specific levels of context. This approach requires the proper configuration of VNs’ topologies and resources, and the definition of local and global (distributed) mechanisms to reconfigure VNs that best fit users’ requirements. In this paper, we propose an analytical model to evaluate the impact of network virtualization and the complexity of the discovery and extension mechanisms defined for VN reconfiguration. Through a delay-based approach, we show the effectiveness of the architecture to deal with different communication requirements and with distinct scenarios for user connectivity establishment, even in the presence of user mobility or using a real WMN topology. The analytical model is compared against a simulation one, showing similar results.

Network Virtualization: Breaking the Performance Barrier

The recent resurgence in popularity of virtualization has led to its use in a growing number of contexts, many of which require high-performance networking. Consider server consolidation, for example. The efficiency of network virtualization directly impacts the number of network servers that can effectively be consolidated onto a single physical machine. Unfortunately, modern network virtualization techniques incur significant overhead, which limits the achievable network performance. We need new network virtualization techniques to realize the full benefits of virtualization in network-intensive domains.

WOW: Self-organizing Wide Area Overlay Networks of Virtual Workstations

This paper describes WOW, a distributed system that combines virtual machine, overlay networking and peer-to-peer techniques to create scalable wide-area networks of virtual workstations for high-throughput computing. The system is architected to: facilitate the addition of nodes to a pool of resources through the use of system virtual machines (VMs) and self-organizing virtual network links; to maintain IP connectivity even if VMs migrate across network domains; and to present to end-users and applications an environment that is functionally identical to a local-area network or cluster of workstations. We describe IPOP, a network virtualization technique that builds upon a novel, extensible user-level decentralized technique to discover, establish and maintain overlay links to tunnel IP packets over different transports (including UDP and TCP) and across firewalls. We evaluate latency and bandwidth overheads of IPOP and also time taken for a new node to become fully-routable over the virtual network. We also report on several experiments conducted on a testbed WOW deployment with 118 P2P router nodes over PlanetLab and 33 VMware-based VM nodes distributed across six firewalled domains. Experiments show that the testbed delivers good performance for two unmodified, representative benchmarks drawn from the life-sciences domain. We also demonstrate that the system is capable of seamlessly maintaining connectivity at the virtual IP layer for typical client/server applications (NFS, SSH, PBS) when VMs migrate across a WAN.