A New Benchmark For Evaluation Of Graph-Theoretic Algorithms

TL;DR

This paper introduces a new graph-theoretic benchmark designed to better evaluate the performance of various architectures and programming models on graph algorithms, addressing limitations of existing benchmarks.

Contribution

The authors developed a comprehensive benchmark suite with kernels representing different classes of graph algorithms, capturing their typical run-time behavior accurately.

Findings

Benchmark captures diverse algorithmic behaviors

Designed to be computationally efficient

Addresses shortcomings of previous benchmarks

Abstract

We propose a new graph-theoretic benchmark in this paper. The benchmark is developed to address shortcomings of an existing widely-used graph benchmark. We thoroughly studied a large number of traditional and contemporary graph algorithms reported in the literature to have clear understanding of their algorithmic and run-time characteristics. Based on this study, we designed a suite of kernels, each of which represents a specific class of graph algorithms. The kernels are designed to capture the typical run-time behavior of target algorithms accurately, while limiting computational and spatial overhead to ensure its computation finishes in reasonable time. We expect that the developed benchmark will serve as a much needed tool for evaluating different architectures and programming models to run graph algorithms.