StreamRec: A Recommendation Inference System with CUDA Stream Acceleration

  • Yuean Niu
  • , Zhizhen Xu
  • , Yushu Sun
  • , Chen Xu*
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Deep learning based recommendation models are widely used in various applications. There are often dozens of groups of sparse features in the input of the recommendation model, and each group of features computes the embedding layer independently and applies a separate feature interaction. However, current deep learning frameworks sequentially schedule the execution of all operators into a single CUDA computational stream. Therefore, we propose StreamRec, a stream-based parallel inference system. It assigns the processing of individual features to different CUDA streams for parallel execution. Besides, StreamRec is able to visualize the execution performance and operators assignment results on the web.

Original languageEnglish
Title of host publicationDatabase Systems for Advanced Applications - 29th International Conference, DASFAA 2024, Proceedings
EditorsMakoto Onizuka, Jae-Gil Lee, Yongxin Tong, Chuan Xiao, Yoshiharu Ishikawa, Kejing Lu, Sihem Amer-Yahia, H.V. Jagadish
PublisherSpringer Science and Business Media Deutschland GmbH
Pages480-483
Number of pages4
ISBN (Print)9789819755745
DOIs
StatePublished - 2024
Event29th International Conference on Database Systems for Advanced Applications, DASFAA 2024 - Gifu, Japan
Duration: 2 Jul 20245 Jul 2024

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14856 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference29th International Conference on Database Systems for Advanced Applications, DASFAA 2024
Country/TerritoryJapan
CityGifu
Period2/07/245/07/24

Keywords

  • CUDA stream
  • Recommendation
  • Visualization

Fingerprint

Dive into the research topics of 'StreamRec: A Recommendation Inference System with CUDA Stream Acceleration'. Together they form a unique fingerprint.

Cite this