Deterministic Latency Bounded Network Slice Deployment in IP-Over-WDM Based Metro-Aggregation Networks

As a critical enabler of the next generation networks, network slicing can dynamically and flexibly create virtual networks leveraging necessary network resources for services with different quality of service (QoS) requirements. To satisfy the strict QoS requirements of some vertical industries, e.g., industry automation, the Deterministic Networking (DetNet) concept has been recently proposed to investigate deterministic service provisioning with bounds on service latency, loss, and jitter. A critical issue for providing deterministic services is how to allocate the right amount of resources that ensures the QoS requirements within a network slice. In this paper, we first obtain the amount of network resources needed for a network slice, using the stochastic network calculus (SNC) method, to ensure the end-to-end latency requirement under specific traffic demands. We then study the network slice deployment problem in an IP-over-WDM (wavelength division multiplexing) metro aggregation network and propose a heuristic with three different objectives: minimizing traffic hops, minimizing lightpaths and minimizing wavelengths, which can help network service provider to optimize network deployment under different considerations. The simulation results show the comparison of the resource utilization under different strategies.