A Generic and Highly Efficient Parallel Variant of Boruvka's Algorithm

Abstract

This paper presents (i) a parallel, platformindependent variant of Borůvka's algorithm, an efficient Minimum Spanning Tree (MST) solver, and (ii) a comprehensive comparison of MST-solver implementations, both on multi-core CPU-chips and GPUs. The core of our variant is an effective and explicit contraction of the graph. Our multi-core CPU implementation scales linearly up to 8 threads, whereas the GPU implementation performs considerably better than the optimal number of threads running on the CPU. We also show that our implementations outperform all other parallel MST-solver implementations in (ii), for a broad set of publicly available roadnetwork graphs.

The analysis and application of Prim algorithm, Kruskal algorithm, Boruvka algorithm

“…The Boruvka Algorithm is an old algorithm. When Otakar Boruvka put forward the algorithm in 1926 [5], the modern general-purpose computer had not even been invented. But with the development of computer science and mathematics, people found that this algorithm had some unique advantages.…”