arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1504.04804cs.DC
收藏

多GPU图形分析

Multi-GPU Graph Analytics

Yuechao Pan, Yangzihao Wang, Yuduo Wu, Carl Yang, John D. Owens

摘要:

我们提出了一个单节点,多GPU可编程图形处理库,允许程序员轻松地扩展单GPU图形算法,以实现可扩展的性能与数十亿条边的大型图形。直接使用单GPU实现,我们的设计只需要程序员指定一些算法相关的问题,隐藏大多数多GPU相关的实现细节。我们分析了在不同的图形原语和数据集的背景下,可扩展性的理论和实践限制。我们描述了几个优化,如方向优化遍历,和一个刚刚足够的内存分配方案,更好的性能和更小的内存消耗。与以前的工作相比,我们在操作和数据集上实现了一流的性能,包括随着系统中GPU数量的增加,在大多数原语上具有出色的强可扩展性和弱可扩展性。

英文摘要:

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 primitives 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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑