arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2207.00514cs.DC
收藏

一种在 GPU 上计算欧几里得最小生成树的单树算法

A single-tree algorithm to compute the Euclidean minimum spanning tree on GPUs

Andrey Prokopenko, Piyush Sao, Damien Lebrun-Grandié

AI总结:

本文提出了一种基于 Borůvka 的单树 GPU 算法来计算欧几里得最小生成树,通过高效最近邻算法减少距离计算,在多种硬件上展现出可移植性与可扩展性,并实现 4-24 倍加速。

摘要:

计算欧几里得最小生成树(EMST)是许多算法中计算量繁重的一步。虽然已知存在工作高效的串行和多线程算法来计算 EMST,但由于复杂的分支结构、数据依赖性和负载不平衡,设计高效的 GPU 算法具有挑战性。在本文中,我们提出了一种基于 Borůvka 的单树算法,用于在 GPU 上计算 EMST。我们使用高效的最近邻算法,并通过避免遍历具有相同组件中叶节点的子树来减少所需的距离计算次数。所开发的算法使用 ArborX(一个基于 Kokkos 框架的开源几何搜索库)以性能可移植的方式实现。我们在各种 2D 和 3D 数据集上评估了所提出的算法,并将其与当前最先进的开源 CPU 实现进行了展示和比较。我们证明了相对于最快的多线程实现取得了 4-24 倍的加速。我们通过在各种硬件上提供结果来证明我们实现的移植性:AMD EPYC 7763、Nvidia A100 和 AMD MI250X。我们展示了该实现的可扩展性,在单个 A100 Nvidia GPU 上不到 0.5 秒的时间内计算出 3700 万个 3D 宇宙学数据集的 EMST。

英文摘要:

Computing the Euclidean minimum spanning tree (EMST) is a computationally demanding step of many algorithms. While work-efficient serial and multithreaded algorithms for computing EMST are known, designing an efficient GPU algorithm is challenging due to a complex branching structure, data dependencies, and load imbalances. In this paper, we propose a single-tree Borůvka-based algorithm for computing EMST on GPUs. We use an efficient nearest neighbor algorithm and reduce the number of the required distance calculations by avoiding traversing subtrees with leaf nodes in the same component. The developed algorithms are implemented in a performance portable way using ArborX, an open-source geometric search library based on the Kokkos framework. We evaluate the proposed algorithm on various 2D and 3D datasets, show and compare it with the current state-of-the-art open-source CPU implementations. We demonstrate 4-24x speedup over the fastest multi-threaded implementation. We prove the portability of our implementation by providing results on a variety of hardware: AMD EPYC 7763, Nvidia A100 and AMD MI250X. We show scalability of the implementation, computing EMST for 37 million 3D cosmological dataset in under a 0.5 second on a single A100 Nvidia GPU.

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。

↑