Accelerating Virtual Network Embedding with Graph Neural Networks

Habibi, Farzad; Dolati, Mahdi; Khonsari, Ahmad; Ghaderi, Majid

doi:10.23919/cnsm50824.2020.9269128

Cited by 19 publications

(16 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• GVine is a state-of-the-art learning-based VNM algorithm, GraphVine [30]. This algorithm uses a generative adversarial network model to perform substrate network embedding and K-means-based clustering to aggregate virtual nodes, thereby establishing an efficient node mapping process.…”

Section: Discussionmentioning

confidence: 99%

“…Because of the structural similarity between graph data and communication network topologies where nodes and edges exist, graph neural networks are used to adapt deep learning techniques for VNM problems. In [30], GraphVine exploited a GCNbased autoencoder to represent complex graph data in a latent space and cluster similar nodes in a substrate network through the latent space. A cluster is selected for assigning the nodes of a selected virtual network.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Gemma: Reinforcement Learning-Based Graph Embedding and Mapping for Virtual Network Applications

et al. 2021

View full text Add to dashboard Cite

Virtual network mapping (VNM) is a challenge in the field of network virtualization. As VNM variants have been formalized depending on substrate network structures, virtual network specifications, mapping optimization objectives, and other factors, a number of VNM heuristic methods have been introduced. On the other hand, reinforcement learning (RL) algorithms have been incorporated into deep learning frameworks and recognized as a promising solution for solving complex resource allocation problems. In this paper, we present an RL-based graph embedding and mapping framework, Gemma, for tackling various VNM problems in a unified end-to-end manner. In the framework, we employ an encoder-decoder deep learning architecture and propose several optimization schemes such as two-stage mapping and model-based selective embedding. Aiming to deal with large-scale VNM problems in both online and offline scheduling systems, the proposed schemes explore the trade-off between inference accuracy and mapping function runtimes, enhancing scalability and timeliness. Gemma shows robust performance under various problem conditions, outperforming other heuristic and learning-based methods.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Gemma: Reinforcement Learning-Based Graph Embedding and Mapping for Virtual Network Applications

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Some recent papers [20][21] process the problem with a deep neural network for performing the embedding (note that in this section we do not consider approaches using neural networks in conjunction with RL). In [20], the graph is clustered with a graph neural network, which then helps guide the embedding procedure. On the other hand [21] pre-processes the network in order to reduce the state-space, making the problem more manageable for other algorithms.…”

Section: Graph Neural Networkmentioning

confidence: 99%

“…Overall, [21] addresses a slightly different problem than we do, since the paper is concerned with feeding a VNE algorithm (such as ours) with hints for solving the VNE, and both could be used in conjunction. On the other hand, [20] is concerned with the VNE, and although it has good results, the runtime is a significant problem as it is exponential in the number of nodes. The authors patch this issue with the use of a GPU.…”

Section: Graph Neural Networkmentioning

confidence: 99%

Monkey Business: Reinforcement learning meets neighborhood search for Virtual Network Embedding

Elkael¹,

Aba²,

Araldo³

et al. 2022

Preprint

View full text Add to dashboard Cite

In this article, we consider the Virtual Network Embedding (VNE) problem for 5G networks slicing. This problem requires to allocate multiple Virtual Networks (VN) on a substrate virtualized physical network while maximizing among others, resource utilization, maximum number of placed VNs and network operator's benefit. We solve the online version of the problem where slices arrive over time. Inspired by the Nested Rollout Policy Adaptation (NRPA) algorithm, a variant of the well known Monte Carlo Tree Search (MCTS) that learns how to perform good simulations over time, we propose a new algorithm that we call Neighborhood Enhanced Policy Adaptation (NEPA). The key feature of our algorithm is to observe NRPA cannot exploit knowledge acquired in one branch of the state tree for another one which starts differently. NEPA learns by combining NRPA with Neighbordhood Search in a frugal manner which improves only promising solutions while keeping the running time low. We call this technique a monkey business because it comes down to jumping from one interesting branch to the other, similar to how monkeys jump from tree to tree instead of going down everytime. NEPA achieves better results in terms of acceptance ratio and revenue-to-cost ratio compared to other state-of-the-art algorithms, both on real and synthetic topologies.

show abstract

“…In the field of ML, VNE algorithm based on graph theory has gradually come into people's view. Habibi et al [30] creatively combined GNN and VNE, and then proposed GraphViNE algorithm based on graph automatic encoder. The biggest feature of this algorithm is that servers with similar resources are clustered, which ultimately effectively reduces the running time of the algorithm.…”

Section: B Vne Solution Based On MLmentioning

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

Accelerating Virtual Network Embedding with Graph Neural Networks

Cited by 19 publications

References 17 publications

Gemma: Reinforcement Learning-Based Graph Embedding and Mapping for Virtual Network Applications

Gemma: Reinforcement Learning-Based Graph Embedding and Mapping for Virtual Network Applications

Monkey Business: Reinforcement learning meets neighborhood search for Virtual Network Embedding

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

Contact Info

Product

Resources

About