The objective of the present study is to compare heuristics of microbioheuristic type employing a scenario with dynamic traffic in the National Science Foundation NETwork (NSFNET) network. Several indicators are used such as the average blocking probability, the usage of the red average, and the usage rate of the red optic. Genetic algorithms, bacterial flagella, and fungal hyphae were compared. The results show that the genetic algorithm has the best response to blocking and that the algorithm based on fungal hyphae has the best use of network resources. The algorithm based on bacterial flagella shows a performance without clear differences, except in the range of 80 to 130 erlangs, where the probability of blocking is much lower. In conclusion, these algorithms can be used for other industrial processes.