Tensor Multiflow Routing Model to Ensure the Guaranteed Quality of Service Based on Load Balancing in Network

Abstract

The article proposes a tensor multiflow routing model with the aim of ensuring the guaranteed Quality of Service based on load balancing in a network. The novelty of this model is that it is focused on providing specified numerical values of Quality of Service indicators such as flow rate and average packet delay. This was achieved by introducing into the multiflow routing model structure the corresponding Quality of Service conditions, which were formulated in an analytical form using the tensor research methodology. At the same time, the space metric was determined by both the service discipline of packets on the routers interfaces and the statistical model of the served flows. In addition, the conditions are formulated in such a way that for each flow transmitted in the network, the resulting average end-to-end packet delays along the set of calculated paths were the same and did not exceed the permissible values. The numerical example has demonstrated the adequacy of the proposed model in terms of the correctness of the obtained calculation results with regard to providing specified values of quality indicators that may differ for each of the packet flows transmitted in the network.