Beyond a Single Queue: Multi-Level-Multi-Queue as an Effective Design for SSSP problems on GPUs

TL;DR

This paper introduces MultiLevelMultiQueue (MLMQ), a GPU data structure for SSSP problems that improves efficiency by reducing synchronization overhead and adapting to diverse inputs, resulting in significant speedups.

Contribution

The paper presents MLMQ, a novel multi-level multi-queue design for GPUs that enhances SSSP performance through adaptive, efficient queue management and inter-queue coordination.

Findings

Achieves 1.87x to 17.13x speedup over state-of-the-art methods.

Introduces a cache-like collaboration mechanism for inter-queue coordination.

Provides an input-adaptive MLMQ configuration scheme.

Abstract

As one of the most fundamental problems in graph processing, the Single-Source Shortest Path (SSSP) problem plays a critical role in numerous application scenarios. However, existing GPU-based solutions remain inefficient, as they typically rely on a single, fixed queue design that incurs severe synchronization overhead, high memory latency, and poor adaptivity to diverse inputs. To address these inefficiencies, we propose MultiLevelMultiQueue (MLMQ), a novel data structure that distributes multiple queues across the GPU's multi-level parallelism and memory hierarchy. To realize MLMQ, we introduce a cache-like collaboration mechanism for efficient inter-queue coordination, and develop a modular queue design based on unified Read and Write primitives. Within this framework, we expand the optimization space by designing a set of GPU-friendly queues, composing them across multiple levels, and further providing an input-adaptive MLMQ configuration scheme. Our MLMQ design achieves average speedups of 1.87x to 17.13x over state-of-the-art implementations. Our code is open-sourced at https://github.com/Leo9660/MLMQ.git.