Exploiting GPUs for fast intersection of large sets

作者：

Highlights：

• Develop hybrid GPU set intersection techniques for set sizes on the order of millions.

• GPU’s fast matrix multiplication greatly improves set intersection for dense datasets.

• GPU-based set intersection techniques are superior over CPU parallel alternatives.

• A co-processing CPU–GPU scheme further improves performance for set containment join.

摘要

•Develop hybrid GPU set intersection techniques for set sizes on the order of millions.•GPU’s fast matrix multiplication greatly improves set intersection for dense datasets.•GPU-based set intersection techniques are superior over CPU parallel alternatives.•A co-processing CPU–GPU scheme further improves performance for set containment join.

论文关键词：Set-intersection,Set-similarity join,Set-containment join,Matrix multiplication,GPU computing,CUDA

论文评审过程：Received 19 June 2021, Revised 3 December 2021, Accepted 28 January 2022, Available online 31 January 2022, Version of Record 12 May 2022.

论文官网地址：https://doi.org/10.1016/j.is.2022.101992