Snowball: A Scalable All-to-All Ising Machine with Dual-Mode Markov Chain Monte Carlo Spin Selection and Asynchronous Spin Updates for Fast Combinatorial Optimization

TL;DR

Snowball is a scalable digital Ising machine that combines dual-mode Markov chain Monte Carlo spin selection with asynchronous updates, significantly reducing solution time for combinatorial optimization problems.

Contribution

It introduces a novel digital architecture supporting all-to-all coupling with configurable precision and dual-mode MCMC, improving convergence speed and scalability.

Findings

Achieved 8x reduction in time-to-solution on benchmark instances.

Supports wide, configurable coupling precision unlike analog counterparts.

Demonstrated effective convergence with asynchronous spin updates.

Abstract

Ising machines have emerged as accelerators for combinatorial optimization. To enable practical deployment, this work aims to reduce time-to-solution by addressing three challenges: (1) hardware topology, (2) spin selection and update algorithms, and (3) scalable coupling-coefficient precision. Restricted topologies require minor embedding; naive parallel updates can oscillate or stall; and limited precision can preclude feasible mappings or degrade solution quality. This work presents Snowball, a digital, scalable, all-to-all coupled Ising machine that integrates dual-mode Markov chain Monte Carlo spin selection with asynchronous spin updates to promote convergence and reduce time-to-solution. The digital architecture supports wide, configurable coupling precision, unlike many analog realizations at high bit widths. A prototype on an AMD Alveo U250 accelerator card achieves an 8$\times$ reduction in time-to-solution relative to a state-of-the-art Ising machine on the same benchmark instance.