QUEST: Multi-Purpose Log-Quantized DNN Inference Engine Stacked on 96-MB 3-D SRAM Using Inductive Coupling Technology in 40-nm CMOS
QUEST: Multi-Purpose Log-Quantized DNN Inference Engine Stacked on 96-MB 3-D SRAM Using Inductive Coupling Technology in 40-nm CMOS
复制标题
DOI：
10.1109/jssc.2018.2871623
复制
发表时间：
2019-01
影响因子：
5.4
通讯作者：
Kodai Ueyoshi;Kota Ando;Kazutoshi Hirose;Shinya Takamaeda-Yamazaki;M. Hamada;T. Kuroda;M. Motomura
中科院分区：
文献类型：
--
作者：
Kodai Ueyoshi;Kota Ando;Kazutoshi Hirose;Shinya Takamaeda-Yamazaki;M. Hamada;T. Kuroda;M. Motomura
Quest是一款可编程多指令、多数据(MIMD)并行加速器，用于通用最先进的深度神经网络(DNN)。它具有三周期延迟的芯片到芯片堆叠、28.8 GB/S、96 MB和使用称为ThruChip接口(TCI)的电感耦合技术的8个SRAM。通过堆叠SRAM而不是DRAM，预计会有更低的内存访问延迟和更简单的硬件。这有助于平衡内存容量、延迟和带宽，所有这些都是尖端DNN在高水平上的需求。Quest还引入了对数量化可编程位精度处理，以在3-D模块中实现更快(更大)的DNN计算(大小)。与线性量化相比，它可以在较低的位宽范围内保持较高的识别精度。Quest芯片的原型集成在40 nm的CMOS工艺中，在二进制精度下达到了7.49万亿次/秒(TOPS)的峰值性能，在300-MHz时钟下达到了1.96TPS的峰值精度。
QUEST is a programmable multiple instruction, multiple data (MIMD) parallel accelerator for general-purpose state-of-the-art deep neural networks (DNNs). It features die-to-die stacking with three-cycle latency, 28.8 GB/s, 96 MB, and eight SRAMs using an inductive coupling technology called the ThruChip interface (TCI). By stacking the SRAMs instead of DRAMs, lower memory access latency and simpler hardware are expected. This facilitates in balancing the memory capacity, latency, and bandwidth, all of which are in demand by cutting-edge DNNs at a high level. QUEST also introduces log-quantized programmable bit-precision processing for achieving faster (larger) DNN computation (size) in a 3-D module. It can sustain higher recognition accuracy at a lower bitwidth region compared to linear quantization. The prototype QUEST chip is integrated in the 40-nm CMOS technology, and it achieves 7.49 tera operations per second (TOPS) peak performance in binary precision, and 1.96 TOPS in 4-bit precision at 300-MHz clock.