Pruned Landmark Labeling Meets Vertex Centric Computation: A Surprisingly Happy Marriage!

TL;DR

This paper introduces a parallel Vertex-Centric implementation of the Pruned Landmark Labeling algorithm, achieving faster performance and scalability on modern multicore architectures while maintaining the same results as the sequential version.

Contribution

It presents a novel VC-PLL algorithm with a batch execution model that reduces computational costs and extends applicability to directed and weighted graphs.

Findings

BVC-PLL runs over twice as fast as the original PLL.

The new algorithm demonstrates high parallel efficiency and scalability.

Theoretical analysis shows lower computational and memory costs than PLL.

Abstract

In this paper, we study how the Pruned Landmark Labeling (PPL) algorithm can be parallelized in a scalable fashion, producing the same results as the sequential algorithm. More specifically, we parallelize using a Vertex-Centric (VC) computational model on a modern SIMD powered multicore architecture. We design a new VC-PLL algorithm that resolves the apparent mismatch between the inherent sequential dependence of the PLL algorithm and the Vertex- Centric (VC) computing model. Furthermore, we introduce a novel batch execution model for VC computation and the BVC-PLL algorithm to reduce the computational inefficiency in VC-PLL. Quite surprisingly, the theoretical analysis reveals that under a reasonable assumption, BVC-PLL has lower computational and memory access costs than PLL and indicates it may run faster than PLL as a sequential algorithm. We also demonstrate how BVC-PLL algorithm can be extended to handle directed graphs and weighted graphs and how it can utilize the hierarchical parallelism on a modern parallel computing architecture. Extensive experiments on real-world graphs not only show the sequential BVC-PLL can run more than two times faster than the original PLL, but also demonstrates its parallel efficiency and scalability.