Improving Parallelism of Breadth First Search (BFS) Algorithm for Accelerated Performance on GPUs

Abstract

Breadth-first search (BFS) is a basis for graph search and a core building block for many higher-level graph analysis applications. However, BFS is also a typical example of parallel computation that is inefficient on GPU architectures. In a graph, a small portion of nodes may have a large number of neighbors, which leads to irregular tasks on GPUs. In addition, the number of nodes in each layer of the graph is also irregular. Therefore, the number of active GPU threads is different for each layer of execution. These irregularities limit the parallelism of BFS executing on GPUs.Unlike the previous works focusing on fine-grained task management to address the irregularity, we propose Virtual-BFS (VBFS) to virtually change the graph itself. By adding virtual vertices, the high-degree nodes in the graph are divided into groups that have an equal number of neighbors, which increases the parallelism such that more GPU threads can work concurrently, and the data set also becomes more regular.Our experimental results show that the VBFS achieves significant speedup over the current GPU implementation of BFS from the Rodinia benchmark [4], and the energy efficiency is also improved.