A Dynamic and Collaborative Multi-Layer Virtual Network Embedding Algorithm in SDN Based on Reinforcement Learning

Lu, Meilian; Gu, Yun; Xie, Dong

doi:10.1109/tnsm.2020.3012588

Cited by 22 publications

(17 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, we propose a heuristic algorithm that can be implemented and evaluated in continuous time. Different from previous dynamic algorithms for wired [29,33] and wireless [34] networks, we consider the resource relationship among changes of different VN requests to improve the embedding profits. In addition, the link interference is considered while mapping virtual nodes and virtual links in wireless airborne field.…”

Section: Brief Summary Of Previous Workmentioning

confidence: 99%

“…On the one hand, some dynamic mapping strategies processed the reconfiguration issues in the assumption that the network status of SN would change as the VN requests existed. In [33], authors proposed a dynamic and collaborative mapping mechanism for a multilayer VNE framework in SDN. If the under VN failed to satisfy the resource demands of up VN, this mechanism adopted a reinforcement learning based approach to increase the resource capacity of virtual elements of under VN or migrating them to other physical nodes or paths of SN.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Dynamic Reliability-Aware Virtual Network Embedding for Airborne Tactical Networks

Miao

Chen

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Airborne tactical networks (ATNs) built on network virtualization (NV) can enable efficient information sharing for network-centric warfare by breaking the tight coupling between applications and network infrastructure and thus solving the network ossification problem. With dynamic changes during military missions, the application of virtualization is challenged by the changing demands on network resources when instantiating multiple virtual networks (VNs) on a shared substrate network (SN), known as virtual network embedding (VNE). However, existing dynamic VNE algorithms, mostly designed for wired networks, focus on the dynamic changes of single VN and do not consider the dynamic changes of different VNs. The proper processing of changes of different VNs is beneficial to improve the embedding profits of VNs. On these backgrounds, we investigate the dynamic wireless VNE that processes the dynamic changes of different VNs in the wireless environment of ATNs. A relationship matrix-based dynamic wireless VNE (DWVNE-RM) algorithm is proposed in order to maximize the acceptance ratio of VN requests. The dynamic changes of different VNs are classified into negative changes (requiring resource) and positive changes (releasing resource). Then, the relationship matrix between negative and positive changes is constructed to properly process the resource relationship among different changes in order to improve the acceptance ratio. In addition, the complex interference of ATNs is considered in the mapping or remapping of virtual nodes and links to improve the reliability. Through extensive simulation, DWVNE-RM is compared with multiple existing dynamic algorithms and outperforms over others.

show abstract

Section: Brief Summary Of Previous Workmentioning

confidence: 99%

mentioning

confidence: 99%

Dynamic Reliability-Aware Virtual Network Embedding for Airborne Tactical Networks

Miao

Chen

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…At the same time, the authors used the method of studying part of physical servers to replace the method of studying all physical servers, which effectively improved the efficiency of the algorithm. In addition, references [35] and [36] are also excellent representatives of dynamic VNE algorithm. But the analysis finds that they do not fully apply ML methods.…”

Section: B Vne Solution Based On MLmentioning

confidence: 99%

“…VNE is a typical NP-hard problem, which determines that the VNE problem can only obtain an approximate optimal solution instead of an optimal solution [10], [11]. Usually, heuristic method is used to solve the VNE process, or VNE is modeled as a combinatorial optimization problem, and then (integer) linear programming is used to solve it.…”

Section: Introductionmentioning

confidence: 99%

Dynamic Virtual Network Embedding Algorithm based on Graph Convolution Neural Network and Reinforcement Learning

Zhang¹,

Wang²,

Kumar³

et al. 2022

Preprint

View full text Add to dashboard Cite

Network virtualization (NV) is a technology with broad application prospects. Virtual network embedding (VNE) is the core orientation of VN, which aims to provide more flexible underlying physical resource allocation for user function requests. The classical VNE problem is usually solved by heuristic method, but this method often limits the flexibility of the algorithm and ignores the time limit. In addition, the partition autonomy of physical domain and the dynamic characteristics of virtual network request (VNR) also increase the difficulty of VNE. This paper proposed a new type of VNE algorithm, which applied reinforcement learning (RL) and graph neural network (GNN) theory to the algorithm, especially the combination of graph convolutional neural network (GCNN) and RL algorithm. Based on a self-defined fitness matrix and fitness value, we set up the objective function of the algorithm implementation, realized an efficient dynamic VNE algorithm, and effectively reduced the degree of resource fragmentation. Finally, we used comparison algorithms to evaluate the proposed method. Simulation experiments verified that the dynamic VNE algorithm based on RL and GCNN has good basic VNE characteristics. By changing the resource attributes of physical network and virtual network, it can be proved that the algorithm has good flexibility.

show abstract

“…The heuristic algorithm [12] used to be the mainstream approach to solve the issue with affordable computation complexity [13,14]. However, along with the emergence of Machine Learning (ML) techniques, such as graph neural networks (GNN) [15,16] and deep reinforcement learning (DRL) [17][18][19][20][21], intelligent algorithms are rapidly used to satisfy the diverse requirements of next-generation wireless networks [22] and design communications system [23,24]. ML approaches can not only simulate wireless communication process with inference [25,26] but also assess resource distribution strategy from a deep perspective due to the nature of extracting information [27].…”

Section: Introductionmentioning

confidence: 99%

Wireless Virtual Network Embedding Algorithm Based on Deep Reinforcement Learning

et al. 2022

View full text Add to dashboard Cite

Wireless network virtualization is widely used to solve the ossification problem of networks, such as 5G and the Internet of Things. The most crucial method of wireless network virtualization is virtual network embedding, which allows virtual networks to share the substrate network resources. However, in wireless networks, link interference is an inherent problem while mapping virtual networks because of the characteristics of wireless channels. To distribute resources efficiently and address the problem of interference, a dynamic embedding algorithm with deep reinforcement learning is proposed. During the training stage, we take resource use and interference from substrate networks as observations to train the agent, and then the agent generates a resource allocation strategy. Aiming at realizing load balance, we reshape the reward function considering the execution ratio and residual ratio of substrate network resources as well as the cost consumed by current virtual network request. Numerical tests show that our embedding approach increases the acceptance ratio and maintains better robustness. Moreover, the results also illustrate that our algorithm maintains a high acceptance ratio while producing less interference and lower cost.

show abstract

A Dynamic and Collaborative Multi-Layer Virtual Network Embedding Algorithm in SDN Based on Reinforcement Learning

Cited by 22 publications

References 24 publications

Dynamic Reliability-Aware Virtual Network Embedding for Airborne Tactical Networks

Dynamic Reliability-Aware Virtual Network Embedding for Airborne Tactical Networks

Dynamic Virtual Network Embedding Algorithm based on Graph Convolution Neural Network and Reinforcement Learning

Wireless Virtual Network Embedding Algorithm Based on Deep Reinforcement Learning

Contact Info

Product

Resources

About