SKYLIGHT: 一种可扩展的百通道3D光子存内张量核心架构用于实时AI推理
SKYLIGHT: A Scalable Hundred-Channel 3D Photonic In-Memory Tensor Core Architecture for Real-time AI Inference
AI总结：
SKYLIGHT是一种可扩展的百通道3D光子存内张量核心架构，通过创新设计实现高效实时AI推理，具备高能效和鲁棒性。
摘要：
人工智能日益增长的计算需求正挑战传统电子器件，使光子计算成为有前景的替代方案。然而，现有光子架构面临根本性的可扩展性和可靠性障碍。本文介绍SKYLIGHT，一种可扩展的3D光子存内张量核心架构，专为实时AI推理设计。通过在3D堆叠中协同设计其拓扑结构、波长路由、累积和编程，SKYLIGHT克服了关键限制。其创新包括低损耗的3D Si/SiN交叉bar拓扑结构、热稳定性的非微环谐振器（MRR）基波分复用（WDM）组件、使用多端口光电探测器（PD）的分层信号累积，以及可编程的非易失性相变材料（PCM）权重。重要的是，SKYLIGHT实现了在位权重更新，支持无标签、层局部学习（例如，前向-前向局部更新），除了推理之外。使用SimPhony进行系统级建模，我们展示了一个单一的144x256 SKYLIGHT核心在单片玻璃上是可行的，并在23.7 TOPS/W下达到342.1 TOPS，使ResNet-50推理在1212 FPS下以27 mJ每图像完成，并在相同工作负载下实现实时测量中的84.17 FPS/W端到端性能（比NVIDIA RTX PRO 6000 Blackwell GPU高1.61倍）。在四个代表性机器学习任务上的系统级评估，包括无监督的局部自学习，展示了SKYLIGHT对现实硬件非理想性的鲁棒性（低比特量化和信号比例的模拟噪声捕获调制、PCM编程和读出变化）。通过噪声感知训练，SKYLIGHT保持了高任务准确率，验证了其作为节能、大规模光子AI加速器综合解决方案的潜力。
英文摘要：
The growing computational demands of artificial intelligence (AI) are challenging conventional electronics, making photonic computing a promising alternative. However, existing photonic architectures face fundamental scalability and reliability barriers. This paper introduces SKYLIGHT, a scalable 3D photonic in-memory tensor core architecture designed for real-time AI inference. By co-designing its topology, wavelength routing, accumulation, and programming in a 3D stack, SKYLIGHT overcomes key limitations. Its innovations include a low-loss 3D Si/SiN crossbar topology, a thermally robust non-micro-ring resonator (MRR)-based wavelength-division multiplexing (WDM) component, a hierarchical signal accumulation using a multi-port photodetector (PD), and optically programmed non-volatile phase-change material (PCM) weights. Importantly, SKYLIGHT enables in-situ weight updates that support label-free, layer-local learning (e.g., forward-forward local updates) in addition to inference. Using SimPhony for system-level modeling, we show that a single 144 x 256 SKYLIGHT core is feasible within a single reticle and delivers 342.1 TOPS at 23.7 TOPS/W, enabling ResNet-50 inference at 1212 FPS with 27 mJ per image, and achieves 84.17 FPS/W end-to-end (1.61 x higher than an NVIDIA RTX PRO 6000 Blackwell GPU) under the same workload in real-time measurements. System-level evaluations on four representative machine learning tasks, including unsupervised local self-learning, demonstrate SKYLIGHT's robustness to realistic hardware non-idealities (low-bit quantization and signal-proportional analog noise capturing modulation, PCM programming, and readout variations). With noise-aware training, SKYLIGHT maintains high task accuracy, validating its potential as a comprehensive solution for energy-efficient, large-scale photonic AI accelerators.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
