HolyLight: A Nanophotonic Accelerator for Deep Learning in Data Centers
Weichen Liu1, Wenyang Liu2, Yichen Ye3, Qian Lou4, Yiyuan Xie3 and Lei Jiang4
1Nanyang Technological University, Singapore
liu@ntu.edu.sg
2Chongqing University, Chongqing, China
3Southwest University, Chongqing, China
4Indiana University Bloomington, USA
ABSTRACT
Convolutional Neural Networks (CNNs) are widely adopted in object recognition, speech processing and machine translation, due to their extremely high inference accuracy. However, it is challenging to compute massive computationally expensive convolutions of deep CNNs on traditional CPUs and GPUs. Emerging Nanophotonic technology has been employed for on-chip data communication, because of its CMOS compatibility, high bandwidth and low power consumption. In this paper, we propose a nanophotonic accelerator, HolyLight, to boost the CNN inference throughput in datacenters. Instead of an all-photonic design, HolyLight performs convolutions by photonic integrated circuits, and process the other operations in CNNs by CMOS circuits for high inference accuracy. We first build HolyLight-M by microdisk-based matrix-vector multipliers. We find analog-todigital converters (ADCs) seriously limit its inference throughput per Watt. We further use microdisk-based adders and shifters to architect HolyLight-A without ADCs. Compared to the stateof-the-art ReRAM-based accelerator, HolyLight-A improves the CNN inference throughput per Watt by 13× with trivial accuracy degradation.
Keywords: Nanophotonic Computing, Convolutional Neural Network, Accelerator.
