Research on Measurement Method of Fine-Grained Traffic in Programmable Network

In order to measure the traffic in the programmable network more accurately, the single-point traffic measurement method and network-level traffic measurement method are studied. In order to overcome the shortcomings of existing single-point traffic measurement algorithms, a single-point traffic measurement method, Flexible Sketch, based on counter and Sketch, is proposed, the simulation results show that the algorithm can effectively improve the accuracy of the measurement. The key measurement node assignment problem in network-level traffic measurement methods is investigated, and an approximate algorithm based on Lagrangian relaxation is proposed, which can effectively solve for a better measurement node assignment scheme. Considering the positive correlation between the computational complexity and the size of the network, a new genetic algorithm is proposed. The simulation results show that the two algorithms can solve the optimal measurement node allocation scheme, achieve network-level traffic measurement, and obtain higher measurement accuracy.