Pipelining Kruskal's: A Neuromorphic Approach for Minimum Spanning Tree

TL;DR

This paper introduces a neuromorphic, pipelined Kruskal's algorithm for minimum spanning trees that leverages event-driven processing and parallelism, achieving substantial speedups over traditional methods on large graphs.

Contribution

It presents a novel neuromorphic, pipelined approach for MST computation, combining SNN-based sorting and union-find, with significant performance improvements over existing algorithms.

Findings

Achieves speedups of 269.67x to 1283.80x over state-of-the-art methods.

Demonstrates effective decoupling of sorting and union-find stages in neuromorphic hardware.

Shows significant performance advantages in large-scale graph processing.

Abstract

Neuromorphic computing, characterized by its event-driven computation and massive parallelism, is particularly effective for handling data-intensive tasks in low-power environments, such as computing the minimum spanning tree (MST) for large-scale graphs. The introduction of dynamic synaptic modifications provides new design opportunities for neuromorphic algorithms. Building on this foundation, we propose an SNN-based union-sort routine and a pipelined version of Kruskal's algorithm for MST computation. The event-driven nature of our method allows for the concurrent execution of two completely decoupled stages: neuromorphic sorting and union-find. Our approach demonstrates superior performance compared to state-of-the-art Prim 's-based methods on large-scale graphs from the DIMACS10 dataset, achieving speedups by 269.67x to 1283.80x, with a median speedup of 540.76x. We further evaluate the pipelined implementation against two serial variants of Kruskal's algorithm, which rely on neuromorphic sorting and neuromorphic radix sort, showing significant performance advantages in most scenarios.