QNET - Optimizing Network Routing with Quantum Annealers

QNET - Quantum Networks
Example configuration where three individual traffic demands of respectively 50 Gbps from node N2 to node N1, node N1 to node N3, and node N2 to node N3 must be routed optimally.

Continued access to fast internet has become an essential part of life and business and has direct societal impact. There are currently large scale efforts for extending and improving the infrastructure of Germany's wide-area internet networks. An equally important component of these efforts is the ability to identify an optimal routing for an existing network that is time dependent but guarantees optimal access with minimal costs and latency. Constraints such as signal boosting and loss, latency, and transferring speeds must all be taken into account in determining the optimal network configuration. In this project, the D-Wave Advantage Quantum Annealer JUPSI is used to tackle this optimization problem. The investigations have demonstrated that JUPSI provides feasible solutions for optimizing networks with up to 6 nodes. The solutions are equal in quality to a classical approach. From a scaling analysis, it is estimated that real-sized networks, of 12 to 16 nodes, will require a quantum annealer with approximately 50,000 qubits.

Reference: arXiv:2212.07854

Last Modified: 12.05.2023