Efficiently Computing Homomorphic Matches of Hybrid Pattern Queries on Large Graphs

Abstract

In this paper, we address the problem of efficiently finding homomorphic matches for hybrid patterns over large data graphs. Finding matches for patterns in data graphs is of fundamental importance for graph analytics. In hybrid patterns, each edge may correspond either to an edge or a path in the data graph, thus allowing for higher expressiveness and flexibility in query formulation. We introduce the concept of answer graph to compactly represent the query results and exploit computation sharing. We design a holistic bottom-up algorithm called GPM, which greatly reduces the number of intermediate results, leading to significant performance gains. GPM directly processes child constraints in the given query instead of resorting to a post-processing procedure. An extensive experimental evaluation using both real and synthetic datasets shows that our methods evaluate hybrid patterns up to several orders of magnitude faster than existing algorithms and exhibit much better scalability.