On the Efficacy of APUs for Heterogeneous Graph Computation

Accelerated Processing Units (APUs) are central processors that feature integrated GPU cores. In this study, we show that this architecture is well-suited to the domain of graph analytics. Our evaluation shows that a current-generation integrated GPU can out-perform an externally-connected discrete GPU by up to 50 % for the breadth-first search and PageRank algorithms. Furthermore, by operating on data with different characteristics in unison, the CPU and integrated GPU can halve the running time of PageRank on a scale-free dataset. 1