DH-Falcon: A Language for Large-Scale Graph Processing on Distributed Heterogeneous Systems

Unnikrishnan Cheramgalath, Rupesh Nasre, Y. N. Srikant

Abstract: Graph models of social information systems typically contain trillions of edges. Such big graphs cannot beprocessed on a single machine. The graph object must bepartitioned and distributed among machines and processedin parallel on a computer cluster. Programming such systemsis very challenging. In this work, we present DH-Falcon, a graph DSL (domain-specific language) which can be usedto implement parallel algorithms for large-scale graphs, tar-geting Distributed Heterogeneous (CPU and GPU) clusters. DH-Falcon compiler is built on top of the Falcon compiler, which targets single node devices with CPU and multipleGPUs. An important facility provided by DH-Falcon is that itsupports mutation of graph objects, which allows programmerto write dynamic graph algorithms. Experimental evaluationshows that DH-Falcon matches or outperforms state-of-the-art frameworks and gains a speedup of up to 13×.