Complex system simulations can often be represented by an
	evolving graph which evolves with a one-to-one mapping between vertices and
	entities and between edges and communications. Performances depend directly
	on a good load balancing of the entities between available computing
	devices and on the minimization of the impact of the communications between
	them. We use competing colonies of numerical ants, each depositing
	distinctly colored pheromones, to find clusters of highly communicating
	entities. Ants are attracted by communications and their own colored
	pheromones, while repulsion interactions between colonies allow to preserve
	a good distribution.