Abstract
The traditional Internet has encountered a bottleneck in allocating network resources for emerging technology needs. Network virtualization (NV) technology as a future network architecture, the virtual network embedding (VNE) algorithm it supports shows great potential in solving resource allocation problems. Combined with the efficient machine learning (ML) algorithm, a neural network model close to the substrate network environment is constructed to train the reinforcement learning agent. This paper proposes a two-stage VNE algorithm based on deep reinforcement learning (DRL) (TS-DRL-VNE) for the problem that the mapping result of existing heuristic algorithm is easy to converge to the local optimal solution. For the problem that the existing VNE algorithm based on ML often ignores the importance of substrate network representation and training mode, a DRL VNE algorithm based on full attribute matrix (FAM-DRL-VNE) is proposed. In view of the problem that the existing VNE algorithm often ignores the underlying resource changes between virtual network requests, a DRL VNE algorithm based on matrix perturbation theory (MPT-DRL-VNE) is proposed. Experimental results show that the above algorithm is superior to other algorithms.
Highlights
The rapid development of the Internet in the past few decades has made great contributions to the progress of human society [1]
We can draw the conclusion that the reinforcement learning agent learns the relation of substrate network nodes in FAM-deep reinforcement learning (DRL)-virtual network embedding (VNE) algorithm during the training phase, the model can be generalized during the test phase
The above DRL-based VNE algorithm does not consider the dynamic change of resources on substrate network between two VN requests, so we propose the MPT-DRL-VNE algorithm
Summary
The rapid development of the Internet in the past few decades has made great contributions to the progress of human society [1]. The contradiction between the expanding functional requirements of network users and the inefficient and unreasonable allocation of underlying network resources in Internet infrastructure is more significant [4] It is mainly reflected in the following aspects: First, the development of emerging network paradigms such as the IoT and 5G requires strong and robust basic network architecture as support. The proposed algorithm generally has some disadvantages, such as low mapping efficiency, large consumption of resources and not meeting the actual network situation Aiming at these problems, this paper discusses several more perfect VNE algorithms based on DRL [29]. The FAM-DRL-VNE algorithm is proposed to solve the problem of ignoring the importance of substrate network representation and training mode.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
