将64个神经形态芯片集成在晶圆级，实现超低延迟、高能效的类脑计算。
DarwinWafer: A Wafer-Scale Neuromorphic Chip
- 64个芯片封装于300mm硅中介层，用异步晶圆互连替代传统板级连接。
- 每片晶圆含1.5亿神经元和640亿突触，运行功耗约100瓦，能效达4.9皮焦每操作。
- 支持全脑模拟，可高效运行斑马鱼与小鼠大脑模型，适合大规模神经计算研究。
类脑计算有望实现类似人脑的能效，但当前多芯片系统依赖印刷电路板（PCB）连接，导致带宽、延迟和能耗成倍增加，削弱生物算法效率。本文提出达尔文晶圆（DarwinWafer），一种超大规模晶圆级系统，通过在300 mm硅中介层上高密度集成64个Darwin3芯片，用晶圆级集成取代片外互联。每个芯片内嵌全局异步锁步（GALS）网络与基于事件的异步晶圆结构，并采用分层时间步同步机制，实现晶圆级低延迟协同运行。每个芯片含235万神经元和1亿突触，整片晶圆共实现1.5亿神经元和640亿突触。在333 MHz、0.8 V下，功耗约100 W，能效达4.9 pJ/SOP，峰值吞吐量为64 TSOPS（0.64 TSOPS/W）。其成功实现得益于芯片-中介层协同设计流程（包含自研中介层凸块规划器及早期信号/电源完整性与热电闭合分析），以及抗翘曲组装技术，通过PCBlet和柔性探针连接实现可靠、可拆卸的晶圆-板集成。实测显示供电压降仅10 mV，温升均匀（34–36 °C）。应用测试表明，单芯片可模拟两只斑马鱼大脑（斯皮尔曼相关系数r=0.896），32个芯片可映射小鼠大脑（r=0.645）。据我们所知，DarwinWafer首次实现晶圆级神经形态计算，为硅基大规模类脑计算提供了可行且可扩展的技术路径。
原文摘要 · Abstract (English)
Neuromorphic computing promises brain-like efficiency, yet today's multi-chip systems scale over PCBs and incur orders-of-magnitude penalties in bandwidth, latency, and energy, undermining biological algorithms and system efficiency. We present DarwinWafer, a hyperscale system-on-wafer that replaces off-chip interconnects with wafer-scale, high-density integration of 64 Darwin3 chiplets on a 300 mm silicon interposer. A GALS NoC within each chiplet and an AER-based asynchronous wafer fabric with hierarchical time-step synchronization provide low-latency, coherent operation across the wafer. Each chiplet implements 2.35 M neurons and 0.1 B synapses, yielding 0.15 B neurons and 6.4 B synapses per wafer.At 333 MHz and 0.8 V, DarwinWafer consumes ~100 W and achieves 4.9 pJ/SOP, with 64 TSOPS peak throughput (0.64 TSOPS/W). Realization is enabled by a holistic chiplet-interposer co-design flow (including an in-house interposer-bump planner with early SI/PI and electro-thermal closure) and a warpage-tolerant assembly that fans out I/O via PCBlets and compliant pogo-pin connections, enabling robust, demountable wafer-to-board integration. Measurements confirm 10 mV supply droop and a uniform thermal profile (34-36 °C) under ~100 W. Application studies demonstrate whole-brain simulations: two zebrafish brains per chiplet with high connectivity fidelity (Spearman r = 0.896) and a mouse brain mapped across 32 chiplets (r = 0.645). To our knowledge, DarwinWafer represents a pioneering demonstration of wafer-scale neuromorphic computing, establishing a viable and scalable path toward large-scale, brain-like computation on silicon by replacing PCB-level interconnects with high-density, on-wafer integration.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。