is there a clustering algorithm in Spark where the resulting clusters are approximately the same size?

Question

i am running a Spark GraphX pregel algorithm where the vertices are intersection coordinates in latitude/longitude, and edges are road segments - a road network. for argumen