BFS and Coloring-Based Parallel Algorithms for Strongly Connected Components and Related Problems

Abstract

Finding the strongly connected components (SCCs) of a directed graph is a fundamental graph-theoretic problem. Tarjan's algorithm is an efficient serial algorithm to find SCCs, but relies on the hard-to-parallelize depthfirst search (DFS). We observe that implementations of several parallel SCC detection algorithms show poor parallel performance on modern multicore platforms and large-scale networks. This paper introduces the Multistep method, a new approach that avoids work inefficiencies seen in prior SCC approaches. It does not rely on DFS, but instead uses a combination of breadth-first search (BFS) and a parallel graph coloring routine. We show that the Multistep method scales well on several real-world graphs, with performance fairly independent of topological properties such as the size of the largest SCC and the total number of SCCs. On a 16-core Intel Xeon platform, our algorithm achieves a 20× speedup over the serial approach on a 2 billion edge graph, fully decomposing it in under two seconds. For our collection of test networks, we observe that the Multistep method is 1.92× faster (mean speedup) than the state-of-the-art Hong et al. SCC method. In addition, we modify the Multistep method to find connected and weakly connected components, as well as introduce a novel algorithm for determining articulation vertices of biconnected components. These approaches all utilize the same underlying BFS and coloring routines.

A Multi-Threading Algorithm to Detect and Remove Cycles in Vertex- and Arc-Weighted Digraph

“…If the graph is not sparse, line 2 of Algorithm 1 can be deleted. Moreover, commodity PCs cannot solve the problem, even use of MPIWS in case of the graph is too large, so parallel algorithms depending on computer cluster must be applied, such as the algorithms proposed in [14,21,22].…”

Modern Platform for Parallel Algorithms Testing: Java on Intel Xeon Phi

“…Another example, performance of graph algorithms proposed by G.M. Slota, was verified with 16 cores CPU [5]. It is clear, that with the increasing number of concurrent threads, the speedup tests become much more reliable.…”

A Novel Edge Cover based Graph Coloring Algorithm

“…There are number of algorithm like, Cuckoo optimisation algorithm [3], modified cuckoo optimisation algorithm [4], polynomial 3-SAT encoding algorithm [5], Ant colony optimisation algorithm [6], Mimetic algorithm [7], GA with multipoint guided mutation algorithm [8] many more are sequential graph colouring algorithm. On the other hand Parallel largest-log-degree-first (LLF) [9], Parallel smallest-log-degree-first (SLF) [9], a parallel algorithm based on BRS [10], parallel graph colouring on multi core CPUs [11] are a parallel algorithm. The parallel algorithm is more time efficient then sequential algorithm due to parallel execution of different iterations of the algorithm.…”