Multi-GPU Graph Analytics [article]

Yuechao Pan, Yangzihao Wang, Yuduo Wu, Carl Yang, John D. Owens
2017 arXiv   pre-print
We present a single-node, multi-GPU programmable graph processing library that allows programmers to easily extend single-GPU graph algorithms to achieve scalable performance on large graphs with billions of edges. Directly using the single-GPU implementations, our design only requires programmers to specify a few algorithm-dependent concerns, hiding most multi-GPU related implementation details. We analyze the theoretical and practical limits to scalability in the context of varying graph
more » ... tives and datasets. We describe several optimizations, such as direction optimizing traversal, and a just-enough memory allocation scheme, for better performance and smaller memory consumption. Compared to previous work, we achieve best-of-class performance across operations and datasets, including excellent strong and weak scalability on most primitives as we increase the number of GPUs in the system.
arXiv:1504.04804v4 fatcat:o2g3vpa6andmnfh33uwu4t3nle