TAXI: Traveling Salesman Problem Accelerator with X-bar-based Ising Macros Powered by SOT-MRAMs and Hierarchical Clustering

2025-04-17

AI Summary

Existing hierarchical-clustering-based Ising solvers for ultra-large-scale Traveling Salesman Problems (TSPs)—up to 85,900 cities—suffer from two critical bottlenecks: rapidly deteriorating solution quality with increasing problem scale and low hardware resource utilization. Method: This paper proposes TAXI, an in-memory computing accelerator featuring hardware–algorithm co-design: (1) X-bar-structured Ising macro-units enabling fully parallel subproblem solving; (2) spin-orbit torque MRAM (SOT-MRAM) as a compact, low-overhead, intrinsic stochastic annealing source; and (3) hardware-aware hierarchical clustering and Ising mapping optimization. Contribution/Results: On the 85,900-city TSP instance, TAXI achieves solutions within 20% of the Concorde optimal tour length—significantly outperforming prior clustering-based Ising solvers—while delivering an average 8× speedup, alongside substantial reductions in energy consumption and latency.

Technology Category

• Machine Learning: Hardware-aware ML
• Search and Optimization: Sampling/Simulation-based Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Ising solvers with hierarchical clustering have shown promise for large-scale Traveling Salesman Problems (TSPs), in terms of latency and energy. However, most of these methods still face unacceptable quality degradation as the problem size increases beyond a certain extent. Additionally, their hardware-agnostic adoptions limit their ability to fully exploit available hardware resources. In this work, we introduce TAXI -- an in-memory computing-based TSP accelerator with crossbar(Xbar)-based Ising macros. Each macro independently solves a TSP sub-problem, obtained by hierarchical clustering, without the need for any off-macro data movement, leading to massive parallelism. Within the macro, Spin-Orbit-Torque (SOT) devices serve as compact energy-efficient random number generators enabling rapid"natural annealing". By leveraging hardware-algorithm co-design, TAXI offers improvements in solution quality, speed, and energy-efficiency on TSPs up to 85,900 cities (the largest TSPLIB instance). TAXI produces solutions that are only 22% and 20% longer than the Concorde solver's exact solution on 33,810 and 85,900 city TSPs, respectively. TAXI outperforms a current state-of-the-art clustering-based Ising solver, being 8x faster on average across 20 benchmark problems from TSPLib.

Problem

• Improves solution quality for large-scale TSPs
• Enables hardware-algorithm co-design for efficiency
• Reduces latency and energy in Ising solvers

Innovation

• X-bar-based Ising macros for parallel TSP solving
• SOT-MRAMs enable efficient natural annealing
• Hardware-algorithm co-design boosts performance