We develop and extensively evaluate highly scalable distributed-memory algorithms for computing minimum spanning trees (MSTs). At the heart of our solutions is a scalable variant of Boruvka's algorithm. For partitioned graphs with many local edges, we improve this with an effective form of contracting local parts of the graph during a preprocessing step. We also adapt the filtering concept of the best practical sequential algorithm to develop a massively parallel Filter-Boruvka algorithm that is very useful for graphs with poor locality and high average degree. Our experiments indicate that our algorithms scale well up to at least 65 536 cores and are up to 800 times faster than previous distributed MST algorithms.

翻译：我们开发并广泛评价了高可扩缩的分布式模拟算法,用于计算最小的横贯树木。我们解决方案的核心是波鲁夫卡算法的可扩缩变方。对于含有许多本地边缘的分隔图,我们通过在预处理步骤期间将图的局部部分承包的一种有效形式改进了这一算法。我们还调整了最佳实际序列算法的过滤概念,以开发一种大规模平行的过滤式波鲁夫卡算法,这对位置差、平均程度高的图表非常有用。我们的实验表明,我们的算法规模至少达到65,536个核心,比先前分布的MST算法快800倍。