Theoretically Efficient Parallel Graph Algorithms Can Be Fast and Scalable

Abstract

There has been significant recent interest in parallel graph processing due to the need to quickly analyze the large graphs available today. Many graph codes have been designed for distributed memory or external memory. However, today even the largest publicly-available real-world graph (the Hyperlink Web graph with over 3.5 billion vertices and 128 billion edges) can fit in the memory of a single commodity multicore server. Nevertheless, most experimental work in the literature report results on much smaller graphs, and the ones for the Hyperlink graph use distributed or external memory. Therefore, it is natural to ask whether we can efficiently solve a broad class of graph problems on this graph in memory. This paper shows that theoretically-efficient parallel graph algorithms can scale to the largest publicly-available graphs using a single machine with a terabyte of RAM, processing them in minutes. We give implementations of theoretically-efficient parallel algorithms for 20 important graph problems. We also present the interfaces, optimizations, and graph processing techniques that we used in our implementations, which were crucial in enabling us to process these large graphs quickly. We show that the running times of our implementations outperform existing state-of-the-art implementations on the largest real-world graphs. For many of the problems that we consider, this is the first time they have been solved on graphs at this scale. We have made the implementations developed in this work publicly-available as the Graph Based Benchmark Suite (GBBS).

Mastiff

“…We evaluate Mastiff in comparison to implementations of Borůvka's algorithm in (1) GBBS [20] (commit 38964eb, OpenMP) and in (2) Galois [43] (release 6). on weight of edges).…”

Engineering Massively Parallel MST Algorithms

“…In the PRAM model, there are multiple algorithms with polylogarithmic span (critical path length of the computation DAG) [1], [13]. More practical algorithms have been devised for the shared-memory (single-node) parallel setting [7], [14]- [16]. The shared-memory Borůvka-variant described in [14] has some similarities to our distributed Filter-Borůvka described in Section V but uses iterative skewed partitioning rather than recursive symmetric partitioning.…”

“…More practical algorithms have been devised for the shared-memory (single-node) parallel setting [7], [14]- [16]. The shared-memory Borůvka-variant described in [14] has some similarities to our distributed Filter-Borůvka described in Section V but uses iterative skewed partitioning rather than recursive symmetric partitioning. We are also not aware of an analysis comparable to ours.…”

“…We are also not aware of an analysis comparable to ours. Very recently, Esfahani et al [17] proposed a structure-aware algorithm that outperforms the previous shared-memory state-of-the-art MST algorithm [14] for graphs with skewed degree distributions. We discuss their results in Section VII.…”

Ray Tracing on a Distributed Computer System Based on Objects of the Internet of Things

“…Сравнение результатов экспериментов с результатами других исследований [7][8][9][10] показывает, что полученные зависимости имеют аналогичный вид и соответствуют зависимостям для обычных параллельных и распределенных вычислительных систем, построенных на основе классических подходов [11,12], для суперкомпьютерных и облачных систем [13,14].…”