F3

Forest Fire (FF) generation model creates graph with heavy-tailed distribution for in- and out-degrees, the Densification Power Law, and shrinking average diameter. These properties of FF find itself important to data scientists because they correspond to real networks evolving over time. As sizes of the networks to be generated are usually large, their memory- and time-efficiencies play important role. However to the best of our knowledge there are no FF generation theoretical and empirical benchmarks presented. Therefore we describe series of experiments on generating graphs via FF model. We expect this paper to provide general comprehension about FF generation complexity and to help choose the optimal library for generating networks, sampling.