已入深夜,您辛苦了!由于当前在线用户较少,发布求助请尽量完整地填写文献信息,科研通机器人24小时在线,伴您度过漫漫科研夜!祝你早点完成任务,早点休息,好梦!

HPPI: A High-Performance Photonic Interconnect Design for Chiplet-Based DNN Accelerators

互连 光子学 计算机科学 计算机体系结构 材料科学 光电子学 电信
作者
Guanglong Li,Yaoyao Ye
出处
期刊:IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems [Institute of Electrical and Electronics Engineers]
卷期号:43 (3): 812-825 被引量:10
标识
DOI:10.1109/tcad.2023.3328828
摘要

In pursuit of higher inference accuracy, the complexity and parameter size of recent deep neural networks (DNNs) have increased significantly. Due to the increasing demand for computing power, the chiplet-based accelerator has been an important computing platform that can handle these DNN models more efficiently. In widely used DNN models, the feature sizes and the number of channels vary greatly among different convolutional layers. Existing chiplet-based accelerators typically adopt consistent optimization strategy for all of the convolutional layers regardless of their sizes, which would limit the inference performance. In this work, we carry out communication-aware customized optimization for convolutional layers with different sizes. First, we propose a reconfigurable high-performance photonic interconnect (HPPI) architecture to facilitate the communication in chiplet-based DNN accelerators. Second, we propose a customized dataflow as the mapping framework and provide four communication patterns of the photonic interconnect with different ways of spatial mapping. Third, we propose a lightweight back propagation neural network to efficiently select the optimal communication pattern for each convolutional layer. The proposed photonic interconnect can be switched between the four communication patterns to enable communication-aware customized optimization for each convolutional layer in the DNN model. As compared to Simba (a representative chiplet-based accelerator with electronic interconnect), HPPI reduces the execution time by 72.23% on average, while saving the energy consumption by 25.49% on average. As compared to ASCEND (a state-of-the-art chiplet-based accelerator with photonic interconnect), HPPI reduces the execution time by 34.04% on average, while saving the energy consumption by 10.34% on average.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
朝阳夕赏完成签到,获得积分10
刚刚
1秒前
顺利的蛋挞完成签到,获得积分10
2秒前
2秒前
2秒前
3秒前
坐雨赏花完成签到 ,获得积分10
3秒前
3秒前
3秒前
星辰大海发布了新的文献求助10
4秒前
蕾蕾发布了新的文献求助30
6秒前
6秒前
外向思松发布了新的文献求助10
7秒前
科研通AI6.2的应助被Tulip采纳,获得10
7秒前
7秒前
钉钉发布了新的文献求助10
7秒前
8秒前
冬雾发布了新的文献求助10
8秒前
8秒前
9秒前
朱帅发布了新的文献求助10
9秒前
10秒前
糊涂的初柔完成签到,获得积分10
10秒前
昆仑发布了新的文献求助10
11秒前
11秒前
12秒前
12秒前
Crssss的应助被文艺的白曼采纳,获得10
12秒前
老实善愁完成签到,获得积分10
12秒前
tutu完成签到,获得积分10
13秒前
学医自救发布了新的文献求助10
14秒前
14秒前
randomname发布了新的文献求助10
14秒前
外向大楚发布了新的文献求助10
15秒前
科研通AI6.3的应助被郭峰采纳,获得30
16秒前
17秒前
英俊的铭的应助被潇洒的代双采纳,获得10
18秒前
18秒前
叹千泠完成签到,获得积分10
19秒前
完美世界的应助被包容耳机采纳,获得30
20秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
Grapes 1000
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8179129
求助须知:如何正确求助?哪些是违规求助? 9569710
关于积分的说明 21806324
捐赠科研通 7774507
什么是DOI,文献DOI怎么找? 3433008
关于科研通互助平台的介绍 2556912
邀请新用户注册赠送积分活动 2478649