RouteNet is a neural architecture for network performance evaluation first proposed in the paper
Unveiling the potential of GNN for network modeling and optimization in SDN by K. Rusek, J. Suárez-Varela, A. Mestres, P. Barlet-Ros, A. Cabellos-Aparicio accepted for ACM Symposium on SDN Research, April 2019, San Jose, CA, USA. arXiv:1901.08113.
An extended version of the model is presented in the paper RouteNet: Leveraging Graph Neural Networks for network modeling and optimization in SDN Krzysztof Rusek, José Suárez-Varela, Paul Almasan, Pere Barlet-Ros, Albert Cabellos-Aparicio arXiv:1910.01508.
If you decide to apply the concepts presented or base on the provided code, please do refer our paper.
@inproceedings{Rusek:2019:UPG:3314148.3314357,
author = {Rusek, Krzysztof and Su\'{a}rez-Varela, Jos{\'e} and Mestres, Albert and Barlet-Ros, Pere and Cabellos-Aparicio, Albert},
title = {Unveiling the Potential of Graph Neural Networks for Network Modeling and Optimization in SDN},
booktitle = {Proceedings of the 2019 ACM Symposium on SDN Research},
series = {SOSR '19},
year = {2019},
isbn = {978-1-4503-6710-3},
location = {San Jose, CA, USA},
pages = {140--151},
numpages = {12},
url = {http://doi.acm.org/10.1145/3314148.3314357},
doi = {10.1145/3314148.3314357},
acmid = {3314357},
publisher = {ACM},
address = {New York, NY, USA},
keywords = {Graph Neural Networks, SDN, network modeling, network optimization},
}
Datasets used for training are available at KDN website
For training simulation, data must be converted to TFrecords using upcdataset.py script.