Exploring the Limits of GPUs With Parallel Graph Algorithms

TL;DR

This paper investigates the performance limits of GPUs for parallel graph algorithms with irregular data access, providing guidelines for adapting PRAM algorithms to improve GPU efficiency in complex graph computations.

Contribution

It offers a detailed analysis of GPU performance for irregular graph algorithms and proposes practical guidelines for adapting PRAM algorithms to GPU architectures.

Findings

PRAM algorithms serve as a good starting point for GPU algorithms

Non-trivial modifications are necessary for efficient GPU implementation

Guidelines are provided for adapting PRAM algorithms to GPUs

Abstract

In this paper, we explore the limits of graphics processors (GPUs) for general purpose parallel computing by studying problems that require highly irregular data access patterns: parallel graph algorithms for list ranking and connected components. Such graph problems represent a worst case scenario for coalescing parallel memory accesses on GPUs which is critical for good GPU performance. Our experimental study indicates that PRAM algorithms are a good starting point for developing efficient parallel GPU methods but require non-trivial modifications to ensure good GPU performance. We present a set of guidelines that help algorithm designers adapt PRAM graph algorithms for parallel GPU computation. We point out that the study of parallel graph algorithms for GPUs is of wider interest for discrete and combinatorial problems in general because many of these problems require similar irregular data access patterns.