Exploiting Computation-Friendly Graph Compression Methods for Adjacency-Matrix Multiplication

Alexandre Francisco, Travis Gagie, Susana Ladra, and Gonzalo Navarro

Computing the product of the adjacency (binary) matrix of a large graph with a real-valued vector is an important operation that lies at the heart of various graph analysis tasks, such as computing PageRank. In this paper we show that some well-known Web and social graph compression formats are computation-friendly, in the sense that they allow boosting the computation. In particular, we show that the format of Boldi and Vigna allows computing the product in time proportional to the compressed graph size. Our experimental results show speedups of at least 2 on graphs that were compressed at least 5 times with respect to the original. We show that other successful graph compression formats enjoy this property as well.