Incidence Constraints in Hypergraph Partitioning on GPU

TL;DR

This paper presents a GPU-accelerated multi-level hypergraph partitioning algorithm that enforces specific constraints, achieving significant speedups and improved partition quality.

Contribution

It introduces a GPU-based hypergraph partitioning method tailored for incidence constraints, with novel algorithms exploiting set sparsity and memory structures.

Findings

Speedups up to 940x over CPU implementations.

Achieved 2-26% better connectivity results.

Effectively enforced size and hyperedge constraints on GPU.

Abstract

Hypergraph partitioning is a pervasive NP-hard problem, and accelerating its computation on GPU can both slice time-to-solution and raise quality of results. In this work, we implement a multi-level hypergraph partitioning algorithm on GPU targeting a specific set of problem constraints: bounded per-partition size and distinct inbound hyperedges. Manipulating hypergraphs requires long orders of nested iterations, and enforcing these constraints introduces further set operations amidst them. Hence, we design algorithms around our problem's specifics, materializing the hypergraph's incidence structure in memory and exploiting set sparsity. Our results show competitive speedups as high as 940x and 2-26% better results in connectivity over a sequential multi-level partitioner.