A 2×30k-Spin Multichip Scalable Annealing Processor Based on a Processing-In-Memory Approach for Solving Large-Scale Combinatorial Optimization Problems

📋 论文概要

本文提出一种基于处理-内存（Processing-In-Memory）方法的多芯片可扩展退火处理器，实现了2×30k自旋（60k自旋）的规模，用于高效求解大规模组合优化问题。该处理器通过多芯片级联扩展自旋数量，并利用内存计算加速退火过程，从而解决NP-hard问题。

💡 主要创新点

• 提出基于处理-内存的多芯片可扩展架构，将退火计算与存储单元紧密集成，减少数据搬移开销。
• 实现2×30k自旋的大规模退火处理器，通过多芯片级联可进一步扩展自旋数量，突破单芯片面积限制。

🏷 关键词

📄 原文摘要

computer architecture, commonly known as annealing processor [1, 2]. An annealing processor provides a fast means for finding the ground state of an Ising model; thus, it can efficiently solve NP-hard combinatorial optimization problems

👥 作者与机构

Takashi Takemoto1, Masato Hayashi2, Chihiro Yoshimura2, Masanao Yamaoka2

Hitachi, Sapporo, Japan Hitachi, Tokyo, Japan 1 2 The last decade has seen impressive progress in the development of a new