English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Network modeling is a critical component for building self-drivingSoftware-Defined Networks, particularly to find optimal routing schemes thatmeet the goals set by administrators. However, existing modeling techniques donot meet the requirements to provide accurate estimations of relevantperformance metrics such as delay and jitter. In this paper we propose a novelGraph Neural Network (GNN) model able to understand the complex relationshipbetween topology, routing and input traffic to produce accurate estimates ofthe per-source/destination pair mean delay and jitter. GNN are tailored tolearn and model information structured as graphs and as a result, our model isable to generalize over arbitrary topologies, routing schemes and variabletraffic intensity. In the paper we show that our model provides accurateestimates of delay and jitter (worst case $R^2=0.86$) when testing againsttopologies, routing and traffic not seen during training. In addition, wepresent the potential of the model for network operation by presenting severaluse-cases that show its effective use in per-source/destination pairdelay/jitter routing optimization and its generalization capabilities byreasoning in topologies and routing schemes not seen during training.

Unveiling the potential of Graph Neural Networks for network modeling and optimization in SDN