LightML: A Photonic Accelerator for Efficient General Purpose Machine Learning
Liang Liu, Sadra Rahimi Kari, Xin Xin, Nathan Youngblood, Youtao Zhang, Jun Yang
摘要
The rapid integration of AI technologies into everyday life across sectors such as healthcare, autonomous driving, and smart home applications requires extensive computational resources, placing strain on server infrastructure and incurring significant costs.We present LightML, the first system-level photonic crossbar design, optimized for high-performance machine learning applications.This work provides the first complete memory and buffer architecture carefully designed to support the high-speed photonic crossbar, achieving over 80% utilization.LightML also introduces solutions for key ML functions, including large-scale matrix multiplication (MMM), element-wise operations, non-linear functions, and convolutional layers.Delivering 325 TOP/s at only 3 watts, LightML offers significant improvements in speed and power efficiency, making it ideal for both edge devices and dense data center workloads.
问问这篇 Paper
问问你的智能体。
Lune 读过与它相关的顶会 Paper,每个回答都会注明依据哪几篇。
相关 Paper
- Lightwave Fabrics: At-Scale Optical Circuit Switching for Datacenter and Machine Learning SystemsHong Liu, Ryohei Urata, Kevin Yasumura, Xiang Zhou 等SIGCOMM 2023 · 被引用 63 次
- A Mixed-Signal Photonic SRAM-based High-Speed Energy-Efficient Photonic Tensor Core with Novel Electro-Optic ADCMd. Abdullah-Al Kaiser, Sugeet Sunder, Ajey P. Jacob, Akhilesh R. JaiswalDAC 2025 · 被引用 2 次
- PIXEL: Photonic Neural Network AcceleratorKyle Shiflett, Dylan Wright, Avinash Karanth, Ahmed LouriHPCA 2020 · 被引用 56 次
- Lightning: A Reconfigurable Photonic-Electronic SmartNIC for Fast and Energy-Efficient InferenceZhizhen Zhong, Mingran Yang, Jay Lang, Christian Williams 等SIGCOMM 2023 · 被引用 25 次
- CrossLight: A Cross-Layer Optimized Silicon Photonic Neural Network AcceleratorFebin Sunny, Asif Mirza, Mahdi Nikdast, Sudeep PasrichaDAC 2021 · 被引用 92 次
