Alleviating All-to-All Communication for Deep Learning Recommendation Model Inference

Author:

Huang Songjun1ORCID,Li Yihong1ORCID,Chen Liangkun1ORCID,Zhang Xiaoxi1ORCID,Liu Shuo2ORCID,Duan Jingpu3ORCID,Wu Wenfei4ORCID,Chen Xu1ORCID

Affiliation:

1. Sun Yat-sen University, China

2. Huawei, China

3. Peng Cheng Laboratory, China

4. Peking University, China

Funder

NSFC

Publisher

ACM

Reference5 articles.

1. Aditya Desai, Li Chou, and Anshumali Shrivastava. 2022. Random Offset Block Embedding (ROBE) for compressed embedding tables in deep learning recommendation systems. In Conference on Machine Learning and Systems.

2. Mixed Dimension Embeddings with Application to Memory-Efficient Recommendation Systems

3. Software-Hardware Co-design of Heterogeneous SmartNIC System for Recommendation Models Inference and Training

4. Semantic-Aware Lossless Data Compression for Deep Learning Recommendation Model (DLRM)

5. Cheetah: Accelerating Database Queries with Switch Pruning

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3