184QPS/W 64Mb/mm<sup>2</sup>3D Logic-to-DRAM Hybrid Bonding with Process-Near-Memory Engine for Recommendation System-Reference-Cited by-同舟云学术

184QPS/W 64Mb/mm²3D Logic-to-DRAM Hybrid Bonding with Process-Near-Memory Engine for Recommendation System

Published:2022-02-20 Issue: Volume: Page:
ISSN:
Container-title:2022 IEEE International Solid- State Circuits Conference (ISSCC)
language:
Short-container-title:

Author:

Niu Dimin¹,Li Shuangchen¹,Wang Yuhao¹,Han Wei¹,Zhang Zhe²,Guan Yijin²,Guan Tianchan³,Sun Fei¹,Xue Fei¹,Duan Lide¹,Fang Yuanwei¹,Zheng Hongzhong¹,Jiang Xiping⁴,Wang Song⁴,Zuo Fengguo⁴,Wang Yubing⁴,Yu Bing⁴,Ren Qiwei⁴,Xie Yuan¹

Affiliation:

1. Alibaba DAMO Academy,Sunnyvale,CA

2. Alibaba DAMO Academy,Beijing,China

3. Alibaba DAMO Academy,Shanghai,China

4. UnilC,Xian,China

Publisher

IEEE

Link

http://xplorestaging.ieee.org/ielx7/9731529/9731102/09731694.pdf?arnumber=9731694

Reference7 articles.

1. The true Processing In Memory accelerator

2. 25.4 A 20nm 6GB Function-In-Memory DRAM, Based on HBM2 with a 1.2TFLOPS Programmable Computing Unit Using Bank-Level Parallelism, for Machine Learning Applications

3. NVIDIA A100 GPU: Performance & Innovation for GPU Computing

4. 15.1 A Programmable Neural-Network Inference Accelerator Based on Scalable In-Memory Computing

5. A Stacked Embedded DRAM Array for LPDDR4/4X using Hybrid Bonding 3D Integration with 34GB/s/1 Gb 0.88pJ/b Logic-to-Memory Interface;fujun;IEDM,2020

Cited by 14 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Survey of Deep Learning Accelerators for Edge and Emerging Computing;Electronics;2024-07-29

2. In-Memory Computing: Global Energy Consumption, Carbon Footprint, Technology, and Products Status Quo;2024 IEEE 24th International Conference on Nanotechnology (NANO);2024-07-08

3. Exploiting Similarity Opportunities of Emerging Vision AI Models on Hybrid Bonding Architecture;2024 ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA);2024-06-29

4. Communication Minimized Model-Architecture Co-design for Efficient Convolution Acceleration;Proceedings of the Great Lakes Symposium on VLSI 2024;2024-06-12

5. LauWS: Local Adaptive Unstructured Weight Sparsity of Load Balance for DNN in Near-Data Processing;2024 IEEE International Symposium on Circuits and Systems (ISCAS);2024-05-19