Distributed Graph Computation Meets Machine Learning

TuX² is a new distributed graph engine that bridges graph computation and distributed machine learning. TuX² inherits the benefits of elegant graph computation model, efficient graph layout, and balanced parallelism to scale to billion-edge graphs, while extended and optimized for distributed machine learning to support heterogeneity in data model, Stale Synchronous Parallel in scheduling, and a new Mini-batch, Exchange, GlobalSync, and Apply ( MEGA ) model for programming. TuX² further introduces a hybrid vertex-cut graph optimization and supports various consistency models in fault tolerance for machine learning. We have developed a set of representative distributed machine learning algorithms in TuX² , covering both supervised and unsupervised learning. Compared to the implementations on distributed machine learning platforms, writing those algorithms in TuX² takes only about 25 percent of the code: our graph computation model hides the detailed management of data layout, partitioning, and parallelism from developers. The extensive evaluation of TuX² , using large datasets with up to 64 billion of edges, shows that TuX² outperforms PowerGraph/PowerLyra, the state-of-the-art distributed graph engines, by an order of magnitude, while beating two state-of-the-art distributed machine learning systems by at least 60 percent.