Network virtualization over WDM networks

Zhang, Shuqiang; Shi, Lei; Vadrevu, Chaitanya S. K.; Mukherjee, Biswanath

doi:10.1109/ants.2011.6163653

Cited by 18 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each VNR, the number of virtual nodes is uniformly distributed in the range of [2,4], and the virtual nodes are randomly connected with a probability of 50%. The bandwidth requirement of each virtual link is uniformly distributed in the range of [10,100] Gbps with steps of 10 Gbps. The computing resource requirement for each virtual node is random in [1,5] units [11].…”

Section: Simulation Results and Discussionmentioning

confidence: 99%

“…In [8], Peng et al proposed a PLI-aware online VONE algorithm in WDM networks and extended their work for a mixed-line-rate network scenario [9]. Inspired by [5], Zhang et al formulated a mixed ILP model and proposed two heuristic algorithms in WDM networks [10]. In [11], Gong and Zhu addressed the VONE over EONs; spectrum continuity and contiguity were considered in their work, and a layered-graph heuristic algorithm was proposed.…”

Section: Related Workmentioning

confidence: 98%

See 1 more Smart Citation

Dynamic Virtual Network Embedding Over Multilayer Optical Networks

Zhang¹,

Song³

et al. 2015

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

Network virtualization is meant to improve the efficiency of network infrastructure by sharing a physical substrate network among multiple virtual networks. Virtual network embedding (VNE) determines how to map a virtual network request onto a physical substrate. In this paper, we first overview three possible underlying substrates for interdatacenter networks, namely an electrical-layer-based substrate, an optical-layer-based substrate, and a multilayer-based (optical and electrical layer) substrate. Then, the corresponding VNE problems for the three physical substrates are discussed. The work presented focuses on VNE over a multilayer optical network; a key problem is how to map a virtual network request onto either an electrical or optical substrate. We propose an auxiliary graph model to address multilayer virtual network mapping in a dynamic traffic scenario. Different node-mapping and link-mapping policies can be achieved by adjusting weights of the edges of the auxiliary graph, which depends on the purposes of the network operators.

show abstract

Section: Simulation Results and Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 98%

Dynamic Virtual Network Embedding Over Multilayer Optical Networks

Zhang¹,

Song³

et al. 2015

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

show abstract

“…Some studies have investigated the VONE over traditional fixed-grid wavelength division multiplexing (WDM) networks. A mixed integer linear programming (ILP) model was formulated and two greedy heuristic were proposed for offline VONE in WDM networks in [9]. In [10], by considering physical layer impairments (PLIs), algorithms were designed to achieve online PLI-ware VONE in WDM networks.…”

Section: Introductionmentioning

confidence: 99%

Integrated VONE Scheme Over Resource-Virtualized Elastic Optical Networks

et al. 2019

View full text Add to dashboard Cite

To eliminate the ossification of optical infrastructure and adapt to the high-performance Internet applications, optical network virtualization is an important enabler to offer each application type a dedicated virtual optical network (VON). It can provide resource sharing for multiple VON requests on an optical infrastructure, and promise efficient use of network resources. In the paper, we first introduce a new concept of resource-virtualized elastic optical network (RvEON), which not only realizes link-tier spectrum virtualization in each substrate fiber link (SFL), but also node-tier virtualized transponders (vTPs) within each substrate node (SN). The RvEON can be considered as a very promising optical infrastructure for multiple VON requests. We then propose a novel integrated virtual optical network embedding (iVONE) scheme over RvEONs. For the iVONE, the node mapping and link mapping are performed in an integrated way, which is totally different from all existing two-phased VONE schemes. In the RvEONs, a novel integer linear programming (ILP) model is formulated, which additionally takes into account the subcarrier and modulator resource constraints of the virtualized TPs. Meanwhile, a novel routing, modulator, spectrum and subcarrier assignment (RMS 2 A) algorithm is designed to establish lightpaths for virtual link mapping in the RvEONs. The simulation results demonstrate that our proposed RMS 2 A algorithm achieves better performance than the existing benchmark two-phased algorithm. We also investigate the impact of bit-rates of the VON requests and the different configurations of various modulators that each vTP can offer on the network performances, respectively. INDEX TERMS Elastic optical networks, network resource virtualization, virtual optical network embedding, virtualized transponder.

show abstract

“…In [8], the effect of substrate path splitting and migration on virtual network embedding were investigated, while in [9] a virtual network mapping algorithm based on subgraph isomorphism detection was proposed that is able to map nodes and links at the same stage. However, the optical network virtualization [10,11] and the coordinated virtualization of both IT and the optical network [12,13] are still at the early stage and have not been extensively studied. Moreover, an optical network has its own specific and rich characteristics, e.g., diverse optical transmission technologies and analogue nature (i.e., constraints and impairments [14]).…”

Section: Introductionmentioning

confidence: 99%

Role of Optical Network Virtualization in Cloud Computing [Invited]

Peng¹,

Nejabati²,

Simeonidou³

2013

J. Opt. Commun. Netw.

View full text Add to dashboard Cite

New and emerging Internet applications are increasingly becoming high-performance and networkbased, relying on optical network and cloud computing services. Due to the accelerated evolution of these applications, the flexibility and efficiency of the underlying optical network infrastructure as well as the cloud computing infrastructure [i.e., data centers (DCs)] become more and more crucial. In order to achieve the required flexibility and efficiency, coordinated provisioning of DCs and optical network interconnecting DCs is essential. In this paper, we address the role of high-performance dynamic optical networks in cloud computing environments. A DC as a service architecture for future cloud computing is proposed. Central to the proposed architecture is the coordinated virtualization of optical network and IT resources of distributed DCs, enabling the composition of virtual infrastructures (VIs). During the composition process of the multiple coexisting but isolated VIs, the unique characteristics of optical networks (e.g., optical layer constraints and impairments) are addressed and taken into account. The proposed VI composition algorithms are evaluated over various network topologies and scenarios. The results provide a set of guidelines for the optical network and DC infrastructure providers to be able to effectively and optimally provision VI services to users and satisfy their requirements.Index Terms-Data center as a service (DCaaS); Optical network virtualization; Physical layer impairments (PLIs); Virtualization infrastructure composition.

show abstract

Network virtualization over WDM networks

Cited by 18 publications

References 4 publications

Dynamic Virtual Network Embedding Over Multilayer Optical Networks

Dynamic Virtual Network Embedding Over Multilayer Optical Networks

Integrated VONE Scheme Over Resource-Virtualized Elastic Optical Networks

Role of Optical Network Virtualization in Cloud Computing [Invited]

Contact Info

Product

Resources

About