基于微环的相干光子GEMM加速器的比较分析
A Comparative Analysis of Microrings Based Incoherent Photonic GEMM Accelerators
AI总结：
本文分析了基于微环谐振器的光子GEMM加速器中四种信号操纵顺序（MASW、ASMW和SMWA）对性能的影响，发现SMWA组织在吞吐量、能效和面积-能效上均显著优于其他方案。
摘要：
已有几种基于微环谐振器（MRR）的模拟光子架构被提出，用于以极高的吞吐量和能效加速深度神经网络中的通用矩阵-矩阵乘法（GEMM）。为实现GEMM功能，这些基于MRR的架构通常以五种不同方式操纵光信号：（i）分离（复制）多个光信号以实现一定的扇出，（ii）聚合（复用）多个光信号以实现一定的扇入，（iii）调制光信号以将输入值印刻到模拟信号幅度上，（iv）对调制后的光信号进行加权以实现模拟输入-权重乘法，（v）光信号求和。基于MRR的GEMM加速器以任意顺序执行前四种信号操纵方式，忽略了这些操纵顺序对其性能的可能影响。本文对具有三种不同操纵顺序的加速器组织进行了详细分析：（1）调制-聚合-分离-加权（MASW），（2）聚合-分离-调制-加权（ASMW），和（3）分离-调制-加权-聚合（SMWA）。我们表明，这些组织以不同程度影响串扰噪声和光信号损耗，这使得这些组织在电路级具有不同级别的处理并行性，在系统级具有不同程度的吞吐量和能效-面积效率。我们对四个CNN模型的评估结果表明，与ASMW和MASW组织平均相比，SMWA组织在吞吐量、能效和面积-能效方面分别实现了高达4.4倍、5倍和5.2倍的提升。
英文摘要：
Several microring resonator (MRR) based analog photonic architectures have been proposed to accelerate general matrix-matrix multiplications (GEMMs) in deep neural networks with exceptional throughput and energy efficiency. To implement GEMM functions, these MRR-based architectures, in general, manipulate optical signals in five different ways: (i) Splitting (copying) of multiple optical signals to achieve a certain fan-out, (ii) Aggregation (multiplexing) of multiple optical signals to achieve a certain fan-in, (iii) Modulation of optical signals to imprint input values onto analog signal amplitude, (iv) Weighting of modulated optical signals to achieve analog input-weight multiplication, (v) Summation of optical signals. The MRR-based GEMM accelerators undertake the first four ways of signal manipulation in an arbitrary order ignoring the possible impact of the order of these manipulations on their performance. In this paper, we conduct a detailed analysis of accelerator organizations with three different orders of these manipulations: (1) Modulation-Aggregation-Splitting-Weighting (MASW), (2) Aggregation-Splitting-Modulation-Weighting (ASMW), and (3) Splitting-Modulation-Weighting-Aggregation (SMWA). We show that these organizations affect the crosstalk noise and optical signal losses in different magnitudes, which renders these organizations with different levels of processing parallelism at the circuit level, and different magnitudes of throughput and energy-area efficiency at the system level. Our evaluation results for four CNN models show that SMWA organization achieves up to 4.4$\times$, 5$\times$, and 5.2$\times$ better throughput, energy efficiency, and area-energy efficiency, respectively, compared to ASMW and MASW organizations on average.
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。
