EUVIMEDEUVIMED entdecken
Crossref

Octopus: Efficient Hypergraph Pattern Mining with Practical Processing-in-Memory Architecture

Yi Zhang, Deting Chen, Yu Huang, Chaoqiang Liu, Haifeng Liu, Jianhui Yue, Xiaofei Liao, Hai Jin

Proceedings of the ACM on Management of Data · 18.05.2026 · 4 (3) · 1-27

DOI: 10.1145/3802094 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the ACM on Management of Data, online erschienen am 18.05.2026. Verfasst von Yi Zhang und 7 weiteren Autor:innen. Die Arbeit nennt 65 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

\noindent Hypergraph pattern mining (HPM) is a key analytical primitive for discovering higher-order relationships in complex data. Despite decades of algorithmic progress, existing systems remain far from saturating available compute resources, as the memory-bound and irregular nature of hypergraph workloads severely constrains sustained throughput. Our empirical characterization shows that even state-of-the-art (SOTA) HPM systems achieve only a small fraction of their theoretical peak performance on real workloads. In this paper, we propose Octopus, the first full-stack hardware-software co-designed system for accelerating HPM on practical processing-in-memory (PIM) hardware. Octopus targets UPMEM, an emerging commercially available PIM platform that integrates thousands of lightweight in-memory compute units within standard DRAM modules. To fully exploit UPMEM's massive parallelism and bandwidth potential while addressing its stringent architectural constraints, Octopus introduces two tightly integrated components: (i) an inter-DPU coordination framework that orchestrates compact data partitioning and balanced workload distribution across thousands of DPUs, and (ii) an intra-DPU mining engine that enables efficient hyperedge-level candidate generation and asynchronous multithreaded execution. We evaluate Octopus on a real UPMEM platform using diverse real-world hypergraph workloads. Experimental results demonstrate up to 55.37×, 19.80×, 1033.89×, and 795.08× speedups over SOTA solutions HGMatch, OHMiner, Pangolin, and PimPam, respectively.

Zitieren

APA

Zhang, Y., Chen, D., Huang, Y., Liu, C., Liu, H., Yue, J., Liao, X., & Jin, H. (2026). Octopus: Efficient Hypergraph Pattern Mining with Practical Processing-in-Memory Architecture. Proceedings of the ACM on Management of Data, 4(3), 1-27. https://doi.org/10.1145/3802094

Vancouver

Zhang Y, Chen D, Huang Y, Liu C, Liu H, Yue J, et al. Octopus: Efficient Hypergraph Pattern Mining with Practical Processing-in-Memory Architecture. Proceedings of the ACM on Management of Data. 2026;4(3):1-27. doi:10.1145/3802094

BibTeX
@article{Zhang2026,
  author = {Zhang, Yi and Chen, Deting and Huang, Yu and Liu, Chaoqiang and Liu, Haifeng and Yue, Jianhui and Liao, Xiaofei and Jin, Hai},
  title = {Octopus: Efficient Hypergraph Pattern Mining with Practical Processing-in-Memory Architecture},
  journal = {Proceedings of the ACM on Management of Data},
  year = {2026},
  volume = {4},
  number = {3},
  pages = {1--27},
  doi = {10.1145/3802094}
}
Kostenlos starten →