Abstract
Graphlet enumeration is a fundamental problem to discover interesting patterns hidden in graphs. It has many applications in science including Biology and Chemistry. In this paper, we present a novel approach to discover these patterns with queries, in a parallel database system. Our solution is based on an efficient partitioning strategy based on randomized vertex coloring, that guarantees perfect load balancing and accurate graphlet enumeration (complete and consistent). To the best of our knowledge, our work is the first to provide an abstract and efficient database solution with queries to enumerate both 3-vertex and 4-vertex patterns on large graphs.