SPACX: Silicon Photonics-based Scalable Chiplet Accelerator for DNN Inference
SPACX: Silicon Photonics-based Scalable Chiplet Accelerator for DNN Inference
复制标题
DOI：
10.1109/hpca53966.2022.00066
复制
发表时间：
2022-04
期刊：
影响因子：
--
通讯作者：
Yuan Li;A. Louri;Avinash Karanth
中科院分区：
文献类型：
--
作者：
Yuan Li;A. Louri;Avinash Karanth
为追求更高的推理精度，深度神经网络（DNN）模型的复杂性和规模显著增加。为克服随之而来的计算难题，人们提出了可扩展的基于小芯片的加速器。然而，在这些基于小芯片的DNN加速器中，使用金属互连的数据通信正成为性能、能效和可扩展性的主要障碍。光子互连由于具有低延迟、高带宽、高能效以及易于广播通信等一些优越特性，能够提供足够的数据通信支持。在本文中，我们提出了SPACX：一种用于DNN推理应用的基于硅光子学的小芯片加速器。具体而言，SPACX包括一个光子网络设计，该设计能够实现无缝的单小芯片和跨小芯片广播通信，以及一种定制的数据流，它促进数据广播并使并行性最大化。此外，我们还探究了光子网络的广播粒度以及其对系统性能和能效的影响。还提出了一种灵活的带宽分配方案，以便为不同类型的数据动态调整通信带宽。使用多个DNN模型的模拟结果表明，与其他最先进的基于小芯片的DNN加速器相比，SPACX能够分别将执行时间和能耗降低78%和75%。
In pursuit of higher inference accuracy, deep neural network (DNN) models have significantly increased in complexity and size. To overcome the consequent computational challenges, scalable chiplet-based accelerators have been proposed. However, data communication using metallic-based interconnects in these chiplet-based DNN accelerators is becoming a primary obstacle to performance, energy efficiency, and scalability. The photonic interconnects can provide adequate data communication support due to some superior properties like low latency, high bandwidth and energy efficiency, and ease of broadcast communication. In this paper, we propose SPACX: a Silicon Photonics-based Chiplet ACcelerator for DNN inference applications. Specifically, SPACX includes a photonic network design that enables seamless single-chiplet and cross-chiplet broadcast communications, and a tailored dataflow that promotes data broadcast and maximizes parallelism. Furthermore, we explore the broadcast granularities of the photonic network and implications on system performance and energy efficiency. A flexible bandwidth allocation scheme is also proposed to dynamically adjust communication bandwidths for different types of data. Simulation results using several DNN models show that SPACX can achieve 78% and 75% reduction in execution time and energy, respectively, as compared to other state-of-the-art chiplet-based DNN accelerators.